Skip to main content

Full text of "Art of the stone age: forty thousand years of rock art"

See other formats


ART OFTH1 EWORLD 



NON-EUROPEAN CULTURES 



THE HISTORICAL, SOCIOLOGICAL 



AND RELIGIOUS BACKGROUNDS 



Digitized by the Internet Archive 

in 2011 with funding from 

LYRASIS Members and Sloan Foundation 



http://www.archive.org/details/artofstoneageforOOunse 



THE ART OF 

THE STONE AGE 

FORTY THOUSAND YEARS OF ROCK ART 

HANS-GEORG BANDI - HENRI BREUIL - LILO BERGER- 

KIRCHNER - HENRI LHOTE - ERIK HOLM 

ANDREAS LOMMEL 




CROWN PUBLISHERS, INC. NEW YORK 



Translated by Ann E. Keep, Dr. Phil. 



The frontispiece represents the head and chest of a 

bull. The horns are in twisted perspective. Above, on 

the left, the head of a horse painted in black and brown. 

Aurignacian-Perigordian. Picture gallery, Lascaux. 




FIRST PUBLISHED IN 1961 
HOLLE AND CO. VERLAG, BADEN-BADEN, GERMANY 

PRINTED IN HOLLAND 

LIBRARY OF CONGRESS CATALOGUE CARD NUMBER 

61-10701 



LIST OF PLATES AND MAPS 



FRANCO-CANTABRIAN ROCK ART 

Lascaux: bull and horse 3 

Altamira: bison 27 

Altamira: hind 31 

Font-de-Gaume: bison 34 

Lascaux: small horses 37 

Niaux: bisons 40 

Niaux: head of a bison 42 

Baume-Latrone: elephant 45 

Le Portel: bisons 47 

Montespan: bear 49 

Gargas: silhouette of hands 51 

Pech-Merle: frieze with horses 54 

Cap Blanc: horse 57 

Rouffignac: head of rhinoceros 59 

Tuc d'Audoubert: bison sculpture 62 

Covalanas: hinds 67 

ROCK ART OF THE SPANISH LEVANT 

Cueva del Charco del Agua Amarga: hunter 74 

Morella la Vella: wild goat 78 

Cuevas del Civil: archer 83 

Cuevas de la Arafia: hind 86 
Cingle de la Mola Remigia: men with bows 89 

Cingle de la Mola Remigia: spider 93 
Cingle de la Mola Remigia: 

anthropomorphous figure 96 

ROCK ART OF THE MAGHREB 
AND SAHARA 

Wadi Djerat: group of rhinoceroses and 

ostriches 104 

Bardai: elephant 107 

Wadi Djerat: ox 110 

Wadi Djerat: heads of pelicans 112 

Wadi Djerat: lion 114 

Bardai: ox with thick horns 117 



Bardai: ox with long horns 
El Greiribat: giraffes 
Wadi Djerat: archer 
Wadi Djerat: fresco 
Wadi Djerat: warriors 
Sefar: masked woman 
Jabbaren: herd of cattle 
Sefar: masked dancers 
Auanrhet: 'White Lady' 
Sefar: shepherd 
Jabbaren: 'Antinea' 

ROCK ART OF SOUTH AFRICA 

Western Transvaal: rhinoceros 
Maretjiesfontein: detail of quagga 
Natal: antelope cows 
Mount St. Paul: 'cloud' elephant 
Silozwane: polychrome giraffes 
Drakensberg Mts.: eland 
Nswatugi: kudu cow 
Leeufontein: rhinoceros and lion 
Western Transvaal: giraffe 
Nswatugi: group of giraffes and zebras 
Nswatugi: giraffe and quagga 
Maretjiesfontein: antelope cow 
Western Transvaal: eland buck 
Drakensberg Mts.: elands resting 
Nswatugi: kudu cows 

ROCK ART OF AUSTRALIA 

Wonalirri: wondjina figure 
Aulen: snake with eagle 
Wonalirri: figures with arcs 
Ngungunda: squatting figure 
Hawker: lizard-like animal 
Ayers Rock: rock painting 



120 
123 
126 

130 
134 
'37 
141 
141 
144 
148 
151 



'55 
160 
163 
166 
169 
172 
174 



■93 

197 



207 
210 

215 
219 
223 

227 



MAPS 



Centres of rock art in South-west Europe 

and North Africa 14 

Distribution of rock art in South-west Europe 18 

Rock art stations in the Dordogne 22 

Rock art stations in the Pyrenees 23 



Rock art stations in the Spanish Levant 72 

Centres of rock art in North Africa 100 

Distribution of rock art in South Africa 157 

Centres of rock art in Australia 208 



SOURCE OF COLOURED PLATES 



Prof. Dr. H.-G. Bandi, Berne 74, 78, 83, 86, 89, 

93.96 

Rene Gardi, Berne 27, 31 

A. Giefers, Pretoria 155, 160, 193 

Jiirgen Harms, Pretoria 163, 166, 172, 181, 197 

Konstanze Holm, Pretoria 169, 174, 178, 185, 187, 

189, 201 

Dr. Henri Lhote, Paris 104, 110, 112, 114, 126, 



130. J34. "37. 141.144. 148.151 
Katharina Lommel, Munich 207, 210, 215, 2ig 
Charles P. Mountford, London 223, 227 

Romain Robert, Tarascon-sur-Ariege 3, 34, 37, 

40, 42, 45, 47, 49, 51, 54, 57, 59, 62, 67 
E. Schulthess (ivith kind permission of Messrs. 

Conzett & Huber, Zurich) 107, 117, 120, 123 



LIST OF FIGURES WITH SOURCES 



3- 
4- 
5- 
6- 

7- 
8- 

9- 



23- 

24- 

=5- 
26- 
27- 
28- 
29- 
30- 

31- 
32- 

33" 
34- 
35" 
36- 
37" 



Hind, Altamira 

Crouching bison, Altamira 

Horse, Les Combarelles 

Lion, Les Combarelles 

Galloping horse, Lascaux 

'Swimming stags', Lascaux 

Small stags, Lascaux 

Cow, Lascaux 

Bison with arrows, Niaux 

Ibex, Niaux 

Trout, Niaux 

Ibexes butting one another, Le Roc de Sers 

Prehistoric elephant, Pindal. After H. Breuil, Four hundred centuries of Cave Art 

Small wild ass, Levanzo 

Unfinished human figure, Valltorta gorge. After H. Obermaier 

Man with bow, Torm6n. After H. Obermaier 

Archer, Cueva Saltadora. After H. Obermaier 

Archer, Cueva del Civil. After H. Obermaier 

Hunter, Cueva de los Caballos. After H. Obermaier 

Archer, Cueva de los Caballos. After H. Obermaier 

Boar hunt, Cueva del Val del Charco del Agua Amarga. After H. Obermaier 

Ibex hunt, Cueva Remigia. After J. B. Porcar, H. Obermaier and H. Breuil 

Hunter, Cueva Remigia. After J. B. Porcar, H. Obermaier and H. Breuil 

Warriors advancing, Cingle de la Mola Remigia. After J. B. Porcar, IH. Obermaier and H. Breuil 

Archers fighting, Morella la Vella. After F. Benitez 

Warrior, Cueva Saltadora. After H. Obermaier 

'Execution', Cueva Remigia. After J. B. Porcar, H. Obermaier and H. Breuil 

Archer with reflex bow, Cueva Vieja. After J. Cabre 

Honey-gatherers, Cuevas de la Arafia. After E. Hernandez-Pacheco 

'Lasso-thrower', Cuevas de la Arafia. After E. Hernandez-Pacheco 

Archer, Mas d'en Josep. After H. Obermaier 

Man with knee-breeches, Els Secans. After H. Obermaier 

Man with bow and cap, Cueva Vieja. After H. Obermaier 

Two figures from the 'Dancing Women', Cogul. After J. Cabre 

Stag hunt, Cueva Vieja. After J. Cabre 

Horse with human figure in double triangle style, Tit 

Recumbent ox, Tarzerouc 



g8 — Figures in double triangle style, Aozon. After Th. Monod 

39 — Rock engraving, Aouilalam 

40 — Recumbent wether, Wadi Amazzar 
411 — War chariot 

42 — Sacred stone with human head 

43 — Recumbent mammal, Silet 

44 — Idol with head of owl, Tabelbalet 

45 —Head of ram, Tamentit 

46 — Schematized chariots, Haut In Daladj 

47 — Goddesses with birds' heads, Jabbaren 

48 — Statuette of girl, Sireuil 

49 — Eland, Vryburg 

50 — Large painting, Philipp Cave near Ameib. After H. Breuil 

51 —Rain myth, La Madeleine 

52 — Rain myth, Djebel Bes Seba. After H. Obermaier and L. Frobenius 

53 — Rain myth, Ain Gudeja. After H. Obermaier and L. Frobenius 

54 — Fish in X-ray style, Oenpelli. After Ch. P. Mountford 

55 —Women running, Oenpelli. After Ch. P. Mountford 

56 — Warriors, Oenpelli. After Ch. P. Mountford 

57 — Warriors, Oenpelli. After Ch. P. Mountford 

58 — Copulation scene, Pilgangura. After H. E. Petri and A. S. Schulz 

59 — Figures resembling human beings, Depuch I. After H. E. Petri and A. S. Schulz 

60 — Hunting scene, Gosford. After D. S. Davidson 

61 — Painting in red and yellow, Walgarock. After D. S. Davidson 

62 — Human and animal figures, Euriowie. After L. Black 

63 — Human and animal figures, Euriowie. After L. Black 

64 — Squatting figure, Conic Range. After D. J. Tugby 

65 — Rock engraving, Devon Downs. After D. S. Davidson 

66 — Rock painting, Malkaia. After D. S. Davidson 



The following have kindly provided the figures in this volume: Konstanze Holm, Pretoria (Figs. 49—53). 
Greta Leuzinger, Zurich (Figs. 1—14, 37, 40, 42—45), Katharina Lommel, Munich (Figs. 54—66), Hannes 
Pixa, Baden-Baden (Figs. 36, 38, 39, 41, 46, 47). 



CONTENTS 

List of plates and maps (5). Source of coloured plates (6). List of figures 
with sources (6). 



11 INTRODUCTION 



15-70 FRANCO-CANTABRIAN ROCK ART 

Discovery (15). Distribution (16). Requirements for preservation (17): 
deposit (19). Age and development (19): Aurignacian-Perigordian (21), 
Solutrian-Magdalenian (24). Implements and technique (25). Origin and 
meaning of Ice Age art (26). Six large caves (28): Altamira (28), Font-de- 
Gaume (32), Les Combarelles (33), Lascaux (36), Niaux (41), Trois Freres 
(43). Caves and rock-shelters in France (45): Chabot, Ebbou, Le Portel 
(46), Tuc d'Audoubert (48), Montespan (48), Gargas (50), Isturitz (53) 
Pech-Merle (53), Sergeac (55), Abri Reverdit, Laussel, Cap Blanc (56), 
La Greze, Barabao (57), Teyjat, Rouffignac (58), Pair -non-Pair (60), Roc 
de Sers (61). Caves in Spain (63): Govalanas, Santian, El Castillo (63), La 
Pasiega, Pindal, Buxu (65), Pena de Candamo, Los Casares (66), La Pileta 
(67). Caves in Italy (68): Levanzo, Romanelli, Addaura (68). 

70-72 FRANCO-CANTABRIAN ART STATIONS 

WITH DATES OF DISCOVERY 



73-98 THE ROCK ART OF THE SPANISH LEVANT .... 

Discovery and study (73). Technique (76). Preservation (77). Style (79): 
expressionism (80). Subject-matter (80). Significance (84). Development 
(88). Dating (88). Distribution (92). Origin (96). 

THE ROCK ART OF THE MAGHREB AND SAHARA . 

99—152 Distribution (99). Discovery (99). Change of climatic conditions (101). 
Dating (103). Technique, tools (105), extinct animal species (108), change 
of climate (109), naturalistic style (111), twisted perspective (112), double 
triangle style (113). Classification (114): bubalus or hunter period, 
pastoralist period, horse period (115), camel period (118). Paintings (1 18): 
sites (119). The Sahara in prehistory (121): tools (121), density of popula- 
tion, race, change of climate (122), agriculture, hunting, 'cattle-raising 
(124). Age of Saharan rock pictures (125): sites in the Wadi Djerat (127), 
pastoralist civilization (128). Chariot period, invasion by the 'people of 
the sea' (129). Chronological coincidence of engraving and painting (132): 



two modes of expression (132), representation of human beings (133; 
Distribution of rock pictures (133): bubalus period (133), period of 'round 
headed men' (136), pastoralist period, origin of the pastoralists (138) 
Composite figures (139). Races (139): expansion of the white race (143) 
Affinities with rock art of Europe and S. Africa (145): stylistic affinities, 
profane art (147). Affinities with Egyptian art (150): chronological dif- 
ferences (150). 



THE ROCK ART OF SOUTH AFRICA 

Means of livelihood .(153). Hunters and food-gatherers (153): sixth sense 
(153), artistic talents, myths (154); prehistory in the living present; the 
Stone Age artist is alive today (156); the problem of dating (156); period 
ization (158); art and artefacts (159); engravings (161); rock galleries (161). 
Area of habitation (162). Regional distribution (164). Significance of sites 
(164). Mythology (167): Mantis (168), legend of the rain (171), totem (173), 
cosmic symbols (174). Examples (175): Nswatugi (176), three giraffes (177), 
constellation (178), ritual destruction, regeneration of the universe (179); 
morphological experience (180); suggestion (182); cosmic significance 
(182). Motivation (184). Drama of cosmic change (186); idols (187); art 
is unhistorical (188); significance for history of art (188); suggestion (190); 
phases (190); technique as criterion (191); colour (192); motifs (196); 
spiritual aspects (202). 



153-204 



THE ROCK ART OF AUSTRALIA 

Rock pictures of Arnhem Land (218). Influences radiating from Oenpelli 
(220). New South Wales (221). Victoria (222). Southern and Central 
Australia (224). Conclusions (226). 



205-231 



APPENDICES 

Bibliography (234). Chronology (2? 



). Glossary (239). Index (242). 



233-249 



INTRODUCTION 



World art does not start with the works of the advanced cultures of 
Mesopotamia and the Nile valley. Its roots lie much further back, in the 
earliest origins of human history. Works of art were already produced by 
prehistoric man, as they are by the primitive peoples that still exist in the 
world today. Fortunately for us many of these early works have been 
preserved up to modern times. In the majority of cases they are rock 
pictures executed by Stone Age hunters who were highly-specialized 
artists. 

It may at first sight seem surprising that the publishers have decided to 
include the Stone Age in the series of volumes on the non-European 
cultures, when the description of the sites given here begins with those 
situated on our own continent. In coming to this decision they were 
guided by the consideration that for two reasons it would have been still 
less fitting to have treated this subject in connection with Western civ- 
ilization: firstly because a great number of finds from other continents 
are also treated in these pages; secondly because the legacy of the 
European Stone Age peoples does not easily fit into the concept of Wes- 
tern culture. 

Furthermore, this volume by no means covers the whole of Stone Age 
art. Only the rock art of this period is dealt with, and a comparatively 
limited section of it at that. The aim here has been to afford readers 
of the Art of the World' series at least a glimpse of the artistic 
achievements of prehistoric peoples in different parts of the world and 
to open up a fascinating field of study which is generally bypassed or 
treated cursorily in other histories of art. 

The study of prehistoric rock pictures is not only a fascinating subject 
but also a dangerous one: fascinating because it gives us an unexpectedly 
rich, realistic and colourful insight into cultures of which otherwise we 
have only the 'skeleton' (mostly stone implements), since everything else 
has vanished long ago; and dangerous because Stone Age rock pictures 
were never 'art for art's sake' but always an expression of certain attitudes 
of mind, and this readily leads to an excessively speculative interpretation. 
The prehistoric art to be discussed in these pages is the product of the 
so-called 'advanced hunter peoples'. Out of the primitive basic type of 
hunter culture, which survived for many millennia and spread over ex- 
tensive areas of the globe, three paths of development gradually evolved: 



one leading to hoeing of the soil; one to cattle-raising; and a third path, 
known as 'advanced hunter culture'. This type of culture, like the orig- 
inal primitive form, definitely still has an acquisitive non-productive 
character. But there is now marked specialization among the hunters, 
bringing with it significant cultural progress. On the other hand, this 
path of development led to a dead end, for in contrast to the two other 
types of economic activity, hoeing of the soil and pasturing cattle, which 
in combination led to settled agriculture, it did not evolve anywhere into 
a fully-developed culture. 

The birth of 'advanced hunter culture' must be sought in the last phases 
of the Ice Age, some 50,000 years ago. Its demise can be observed in our 
own day, for there are still some peoples — the Bushmen in South Africa, 
for example — who are the last surviving practitioners of this once wide- 
spread mode of economic life. 

With peoples in this stage of development their entire existence in all 
aspects — economic, social and spiritual — is dominated by hunting. 
Everything is focussed upon the quest for game. All man's thoughts and 
emotions are centred upon "the constant conflict, practical and spiritual, 
with animals. This leads to a conception of the relationship between man 
and beast as a very real unity of substance, that is to say, to a belief in 
the unity of the entire animal world and the interchangeability of 
human and animal forms of existence." * These far-reaching problems 
cannot be gone into in detail here, but it may be pointed out that in 
many places 'advanced hunters' had to their credit varied artistic 
achievements, a genuine 'ritual art' which gives us most interesting in- 
sights into the spiritual world in which these peoples lived. 
We are concerned here with rock pictures, although the art of the 
'advanced hunters' also comprises minor arts such as sculpture, engraved 
or painted decoration on movable objects, and much else besides. But 
one should not jump to the conclusion that all the rock engravings and 
paintings scattered over the entire globe are the work of 'advanced hunt- 
ers'. Rock pictures can also be produced by peoples in other stages of 
cultural development, although in such cases one must frequently reckon 
with the possibility that influences or traditions derived from hunters 
play a certain role. Naturalistic works are found most particularly in the 
southern parts of the region inhabited by 'advanced hunters'. This fact 
will be illustrated in the various sections of the present volume. These 
sections do not amount to an exhaustive survey. Such a survey would 
have to comprise much more than the few glimpses given here of Franco- 
Cantabrian and Eastern Spanish art in south-western Europe, of the rock 



paintings in North Africa (a region where pictures by hunters are 
interspersed with others by cattle-raisers, which are also treated here), 
of the rock art of South Africa, and finally that of Australia. Some 
other centres of rock pictures, principally in Africa and the Near East, 
have had to be omitted. Nor has it been possible to mention the rock 
art of the Arctic, although it may be pointed out here that in Scandinavia, 
in addition to engravings executed by agriculturists of the Bronze Age, 
there is also a very impressive widely diffused group of works by Stone 
Age hunters and fishermen. Other evidence of Arctic art is to be found 
in Karelia, Western Siberia and Central Asia. Finally, no attention has 
been paid in these pages to rock art in the western hemisphere, which 
also belongs in part to hunter culture. 

Despite these limitations this volume will perhaps help to show that 
the 'ritual art' works of the 'advanced hunters', some of which date back 
to the Ice Age and some to more recent periods, are as impressive and 
beautiful as they are diversified and informative. The fact that the 
various regions of rock art are dealt with independently by different 
authors, and that each is alone responsible for the opinions expressed, 
shows how varied the approach to these problems can be. While some 
authors have kept rather to the traditional method of interpretation others 
have attempted to strike out along new paths. But all endeavour to 
introduce the reader to a subject which, owing to the magnificence of 
many of the works discussed, can also give something to those who are 
not so familiar with the mysterious world in which the minds of these 
hunter peoples moved. 

Hans-Georg Bandi 



1 K. J. Narr, Historia Mundi, Vol. I, Berne, 1952, p. 517 



'3 



THE CHIEF CENTRES OF ROCK ART IN 
SOUTH-WEST EUROPE AND NORTH AFRICA 




FRANCO-CANTABRIAN ROCK ART 

BY 

H. BREUIL AND L. BERGER-KIRCHNER 



At the close of the 19th century the entrances to caves in northern Spain 
and south-western France were rediscovered after they had remained 
sealed for thousands of years. In these caves lay hidden an undreamt-of 
wealth of artistic works dating from the Ice Age: magnificent polychrome 
murals and roof frescoes, decorated partly with figures of animals now 
long extinct, partly with vivid engravings and a vast number of curious 
signs. The disclosure of such an ancient prehistoric art came as a complete 
surprise to a world that had hitherto sought the origins of human artistic 
endeavour in the civilizations of Mesopotamia and Egypt. The discovery 
of Ice Age art had therefore to encounter a multitude of obstacles, and 
it was only a quarter of a century later that this art finally found ac- 
ceptance. 

In 1869 the entrance to the cave of Altamira was discovered by a man 
hunting foxes in a tract of meadow land near the village of Santillana 
del Mar, in northern Spain. Ten years later Marcelino de Sautuola, a 
local nobleman, commenced excavations at Altamira. One day he. took 
his little daughter Maria, then five years old, into the cave. While playing 
in a chamber some 30 yards from the entrance her eye was caught by the 
magnificent frescoes on the roof. Since Sautuola was the first man to have 
entered the cave he was convinced that no modern artist could have 
created these paintings. However, in the scientific world doubts were> 
expressed as to their authenticity. At the Congress of Anthropology and 
Prehistoric Archaeology held at Lisbon in 1880 the Altamira paintings 
were dismissed as forgeries and were soon once again consigned to oblivion. 
In 1895 E. Riviere discovered paintings and engravings in the cave of 
La Mouthe in the Dordogne, and in the following year F. Daleau deci- 
phered several engravings in the Pair-non-Pair cave in the Gironde. These 
pictures, partially covered by thick deposits, were the only ones that 
secured recognition by the experts. 

It was only in the twentieth century, through the work performed by a 
new generation of prehistorians, that Ice Age art came into its own. In 
September 1901 Henri Breuil, then a young man, in collaboration with 



DISCOVERY 



Recognition of 
Ice Age art 



'5 



L. Capitan and D. Peyrony, discovered Les Combarelles near Les Eyzies, 
a cave lavishly decorated with engravings. The very next week the three 
explorers were able to report the discovery of the Font-de-Gaume cave, 
also situated near Les Eyzies in the Vezere valley. The very old sinter 
deposits that covered the engravings at Les Combarelles as well as the 
paintings at Font-de-Gaume gradually dispelled the doubts diat had 
existed hitherto. The similarity between the frescoes of Font-de-Gaume 
and those of Altamira led the venerable prehistorian E. Cartailhac to 
revise his strongly held opinions and to recognize that the pictures orig- 
inated from the Ice Age. With his Mea culpa d'un sceptique the debate 
as to the authenticity of this cave art was shelved. 

In 1903 Peyrony disclosed further engravings at Bernifal and Teyjat 
in the Dordogne. 1906 is a key year: it was then that Alcalde del Rio and 
Padre Sierra discovered the famous caves of El Castillo, Covalanas and 
La Haza, as well as other caves in Cantabria, which were then studied by 
Breuil; in the same year F. Regnault came across ancient Aurignacian 
figures in the cave of Gargas in the French Pyrenees and Molard discovered 
magnificent animal paintings in black in the 'Salon Noir' of the gigantic 
cave at Niaux, which shortly afterwards were examined by Breuil and 
Cartailhac. Not long after this Dr. Jeannel drew attention to the paintings 
at Le Portel (Ariege). It was in 1912 that H. Obermaier and P. Wernert 
penetrated into the cave of La Pasiega, situated not far from Castillo. 
Already before the first world war publications began to appear, in the 
form of magnificent monographs, made possible by the generosity of Prince 
Albert of Monaco. In 1912 and 1914 Count Begouen and his sons disclosed 
near Montesquieu-Avantes (Ariege) the two vast cave complexes of Tuc 
d'Audoubert and Trois Freres. 1940 was the year of tire discovery of 
Lascaux, with its magnificent paintings, which were subsequently studied 
by Breud and F. Windels. The most sensational discovery in the last few 
years was that of the cave of Rouffignac (Dordogne) in the summer of 
1956. As happened a century ago, its paintings touched off a controversy 
with regard to their authenticity; it was only after a long debate among 
the experts that their Ice Age origin was conclusively established. 
It is impossible to discuss within the framework of this volume all the 
numerous art stations discovered since the beginning of the twentieth 
century- An appendix at the end of this chapter, however, provides a 
summary of all the caves at present known for their mural art. 
DISTRIBUTION As the name indicates, Franco-Cantabrian art is fairly narrowly confined 
to certain regions in France and Spain. Apart from one or two offshoots 
three regions have achieved prominence as centres of Ice Age art: the first 

16 



in the south-west o£ France (the Dordogne and adjoining departments); 
the second in the Pyrenean region of southern France; and the third 
in the Gantabrian area of northern Spain. In France the border of 
the Franco-Cantabrian art region is formed in the north-east by caves 
situated in the departments of Vienne and Yonne, and in the east by the 
art stations located in the Ardeche and Gard valleys; there are none 
beyond the Rhone. In Spain the southernmost outposts are to be found 
in Andalusia (La Pileta). Several caves situated in central Spain (Casares) 
can be regarded as links between north and south. Completely isolated, 
on the other hand, are three caves at the extreme southern tip of Italy, 
in Sicily and the Aegadian Islands. Exploration carried out in Belgium, 
Germany and Czechoslovakia, as well as in the Balkans, has not as yet 
produced any notable results. The engraving of a stag found in one cave 
in Bavaria cannot be ascribed to the Palaeolithic age. Traces of painting 
in one cave in England are likewise of uncertain date. 
The paintings and engravings in the Franco-Cantabrian caves were applied 
to the walls, which consist of limestone varying in type from one area to 
another. Only in exceptional cases do other types of rock serve as a surface. 
The question thus arises how these paintings have been preserved and 
what influences — physical, chemical and organic — they were exposed to 
during the course of the millennia. The effects of gravity were particularly 
important: in the innermost parts of the caves, as well as in open rock- 
shelters (abris), the overhanging rocks are continually collapsing, especially 
in stratified lime deposits. The constant changes of temperature are condu- 
cive to fissure of the rock, and this process of weathering is accelerated still 
further by earthquakes. The fallen boulders, decorated in part with paint- 
ings and engravings, are in the course of time covered over by further 
collapses of the roof, by sand and clay carried into the cave by the wind, 
and by the refuse left by human beings. In this way archaeological strata 
were formed. Sculptures embedded in these layers can be preserved so 
long as they are not destroyed by the action of water. Water acts in a 
cave in two ways: by infiltration (ground-water) and by condensation on 
the walls. Condensation is caused by changes of temperature and the 
constant stream of air flowing from the inner parts of the cave to the 
outside world. It attacks the calcareous walls and is one of the chief 
reasons for the destruction of Ice Age mural paintings. Ground-water, on 
the other hand, by percolating through fissures in the rock, causes the 
formation of fantastic bizarre-shaped stalactites and stalagmites hanging 
from the roof or rising up from the floor. In the same way calcareous 
layers are deposited by ground-water on the cave walls, covering the 



CONDITIONS 
REQUIRED FOR 
PRESERVATION 



iV 



DISTRIBUTION OF ROCK ART IN 
SOUTH-WEST EUROPE 

i. Cf. also Map p. 23 

2. Cf. also Map p. 72 

3. Cf. also Map p. 22 

4. Arcy-sur-Cure 

5. Angles-sur-Anglin 

6. Sall61es-Cabard£s 

7. Aldfene 

8. Los Casares 
g. La Cala 

10. Ardales (Cueva Trinidad) 

11. La Pileta 

12. Las Palomas 

13. Levanzo 

14. Addaura 

15. Romanelli 



:© 




© 



© 



©j 



m<£ 




1(10 200 miles 



paintings; this can provide evidence as to their antiquity. This sinter 
deposit also contributes to the preservation o£ the paintings. There are, 
of course, cases where the sinter formation may become so thick that a 
painting or engraving can no longer shimmer through. Even if the depos- 
iting of sinter has ceased and paintings are to be found beneath it, this is a 
guarantee of their antiquity. But if, on the other hand, the depositing 
of sinter is still in progress this merely proves that the paintings to be 
found beneath the deposit do not originate from recent times. At Lascaux 
the walls of the picture gallery are covered with a fine bright sinter 
deposit upon which the pictures were later superimposed. The depositing 
of sinter had already ceased during the Ice Age. 

Cave walls, like all other rock surfaces, are also exposed to biological 
and chemical processes: in lighted parts lichens grow on dry and sometimes 
on wet rock, algae and mosses on rock that is permanently damp. This 
vegetation can often help to destroy the paintings. On a completely dry 
wall oxydization is always found on the surface, causing the natural 
colour of the rock to change, often to a darker colour, and this in turn in 
many places limits the extent to which the paintings are visible. 
Deposit is an important phenomenon to be found in caves. It is of the Deposit 
greatest significance in dating pictures and sometimes in their preserva- 
tion. As has already been mentioned, the floor of the cave is raised up by 
falls of the roof, drifts and human settlement — the latter, however, only 
takes effect at the cave entrance — with the result that in several caves 
extensive deposits form. Walls decorated with paintings are thus partly 
covered over. At several art stations this deposit can also consist merely 
of sand or clay. At Pair-non-Pair deposits reached almost up to the roof; 
walls decorated with engravings disappeared below it, but the paintings 
retained their original brilliance remarkably well. The strata of the 
deposit helped to establish the date of the paintings they had covered 
over. 

Since the discovery of Franco-Cantabrian cave art scholars have been AGE AND 
largely concerned with the question of ascertaining the age of the various 
works. How could the great antiquity of Ice Age art be authenticated? 
What evidence is there that these paintings and engravings do not orig- 
inate from the last few centuries? 

As has already been mentioned, in the course of an immense length of 
time the walls were exposed to the action of water, causing the surface 
to erode or leaving calcareous deposits on the paintings. In addition to 
this the rock was affected by chemical processes, which caused the colour 
of the engraved picture to change, thus indicating great antiquity- 



DEVELOPMENT 



19 



Another characteristic feature of this art is that in most cases the animals 
represented belong to species that are either extinct or are no longer 
to be found in Europe. Pictures of animals such as reindeer, mammoths, 
rhinoceroses, musk-oxen, bisons and saiga antelopes — the latter, however, 
depicted only once, in Combarelles — which inhabited the Franco- 
Cantabrian area during the late Ice Age prove that this form of art, too, 
must be assigned to that era, i.e. that it must originate from the last 
glaciation, which ended about 10,000 B.C. The antiquity of cave art 
is also attested by the fact that the pictures were repeatedly covered by 
strata whose age can be established exactly from the finds embedded in 
them. At Pair-non-Pair the bones of Ice Age animals, as well as remains 
left by humans beings contained in the layers covering the engravings, 
have enabled them to be dated to the Aurignacian period, an early stage 
of cultural development when the first pictures were produced. Works 
of art found below a layer can either be older than that layer or contem- 
poraneous with it, but never more recent. The same applies to boulders 
decorated with paintings which have fallen into the deposit, such as 
have been recovered, for example, from Aurignacian deposits at Sergeac 
in the Dordogne. At Altamira, El Castillo and Gargas small objects of 
art have also come to light in the deposits. These, and also the works of 
mural art, express the creative power of the Ice Age artist. In most cases 
they are reliefs or engravings on horn, antler, bone or small slabs of stone; 
the minor arts, however, also comprise figures sculptured in the round, 
in stone, ivory and baked clay. If such small contour drawings are found 
in the layer this is a guarantee of their antiquity, and stylistic comparisons 
can be drawn with mural paintings. But the different styles correspond 
to different eras and stages of development within Franco-Cantabrian 
cave art. This is in turn attested by the frequent superposition of paintings 
on top of one another. It is clear that the most recent figures are super- 
imposed upon all the others and that the oldest pictures are to be found 
in the deepest strata. Such layers of pictures are to be met with in many 
caves. By means of these, and the criteria mentioned above, Breuil has 
succeeded, in the course of a long life devoted to exploration and study, 
in reconstructing the chronology or sequence in which Ice Age works of 
art appeared. 

Breuil distinguishes between two major periods: Aurignacian-Perigordian 
and Solutrian-Magdalenian. These cycles correspond to the archaeological 
division of the late Ice Age into three main cultures: Aurignacian (after 
the type-site in the Haute Garonne), Perigordian (after the district of 
Perigord), Solutrian (after the type-site of Solutre' in the department of 



Saone-et-Loire), and Magdalenian (after the type-site of La Madeleine in 

the department of the Dordogne). 

It was no't until the beginning of the Upper Palaeolithic, which occurred Aurignacian-Perigordian 

some 30,000 years ago during the last great glaciation in Europe, that the 

first signs of man's artistic endeavours appeared. According to Breuil the 

oldest examples of this art are representations of hands, dating from the 

Middle Aurignacian. Such paintings of hands, which occur at Gargas, 

El Castillo and other caves, are outlined in red and black and stand out 

brightly against their rock backgrounds. The origin of figurative art must, 

however, be seen rather in the lines or incisions drawn on clay with 

several fingers or pronged palette-knives. These 'macaroni', as they are 

called, to be found on many cave walls, which at first are merely a chaotic 

tangle of lines, form the earliest 'pictorial writing'; they probably also 

date from the Middle Aurignacian. In the case of Aurignacian man this 

was probably at first a simple act of emulation, which later evolved 

into ever more complicated forms, until eventually from the jumble of 

meandering lines there developed a purposeful figure: though at first 

still clumsy and amateurish, in the course of time man succeeded in 

creating a simple but distinct image of an animal. Fingers were soon 

replaced by an engraving tool, by means of which lines could also be 

cut into the hard rock. Altamira and Gargas yield fine specimens, in 

which the distinct contours of animal heads suddenly stand out 

from the tangle of lines cut into the clay. These attempts are followed 

by polychrome line drawings of animals no longer executed with the 

fingers but with brush-like implements. From Castillo and Font-de-Gaume 

we have animal pictures painted in this way in red and yellow — though 

rarely in black. They are often found together with simple hut-like 

figures known as tectiforms. Apparently more recent in date are the animal 

figures where the contour is drawn with wide lines; in many cases the 

strokes are jagged, and the bodies of the animals are occasionally outlined 

by a row of closely-spaced dots (Covalanas, Altamira and El Castillo). 

The colour, originally limited to the contour line, now begins to cover 

the body of the animal — at first only certain portions, but soon, as this 

technique develops, the entire body. Representations of this kind, painted 

in red, are to be found at Altamira and Le Portel (Ariege), while black 

and sepia ones may be seen at Font-de-Gaume and Lascaux. Breuil assumes 

that this is the origin of the gradual development of bicolour painting, 

as is suggested by the caves of La Pasiega (Santander) and Pech-Merle 

(Lot). Painted boulders of this kind found in the Aurignacian and 

Perigordian strata of Sergeac confirm that this dating is correct. The 




Rock art areas in the Dordogne 



1. Ebbou 

2. Le Colombier 

3. Le Figuier 

4. Chabot 

5. Bayol 

6. Baume-Latrone 

7. Oullins 

8. Aldene 

9. Salleles-Cabardes 

10. Sainte-Eulalie 

11. Cantal 

12. Pech-Merle 

13. Marcenac 

14. Les Merveilles 

15. Cougnac 

16. La Magdelaine 

17. Gorge d'Enfer 

18. Jean-Blancs 

19. Font-de-Gaume 

20. La Mouthe 

21. La Cal^vie 

22. Bernifal 



23. Beyssac 

24. Commarque 

25. La Greze 

26. Laussel 

27. Lascaux 

28. La Sudrie 

29. Belcayre 

30. Sergeac 

31. Cap Blanc 

32. Les Combarelles 

33. Croze a Gontran 

34. Barabao 

35. La Ferrassie 

36. Rouffignac 

37. Fourneau du Diable 

38. Teyjat 

39. Roc de Sers 

40. Chaire a Calvin 

41. Pair-non-Pair 

42. Le Gabillou 

43. Laugerie Basse 
Laugerie Haute 




Rock art areas in the Pyrenees 



1. 


El Cueto 


2. 


Pena de Candamo 


3- 


Pindal 


4- 


Buxu 


5- 


Mazaculos 


6. 


La Meaza 


7- 
8. 


Quintanal 
Bolado 


9- 

10. 


La Loja 
La Clotilde 


n. 


San Antonio 


12. 


Santian 


!3- 

14. 


Las Aguas 
Altamira 


»5- 


La Haza 


16. 


Covalanas 


17- 


Hornos de la Pena 


18. 


Penches 


!9- 

20. 


Atapuerca 
Los Casares 


21 


El Pendo 



22. El Castillo 
La Pasiega 
Las Monedas 

23. Salitrd 

24. Sottarriza 

25. Santimamifie 

26. Venta de la Perra 

27. Berroberia 

28. Isturitz 

29. Etcheberriko 
go. Labastide 

31. Gargas 

32. Montespan 

33. Marsoulas 

34. Tuc d'Audoubert 

35. Les Trois Freres 

36. Mas d'Azil 

37. Le Portel 

38. Bedeilhac 

39. Ussat 

40. Niaux 



2.3 



large red animal figures at Lascaux with their black and dark brown 
heads must also be assigned to this period. A particularly characteristic 
Fig. 7 feature of Perigordian art at Lascaux is the 'twisted perspective' of the 
horns or antlers of bovidae or cervidae — i.e., these are not represented 
in profile, as is the case with the remainder of the body, but instead are 
viewed from the front. The technique of engraving develops on lines 
parallel to painting in this period. In the first stage drawings are executed 
on wet clay with the fingers; but soon there develop figures of animals 
depicted in a forceful naturalistic style. The legs are still rudimentarily 
suggested and the technique of drawing is still stiff, but all the poten- 
tialities are already at hand that are later to lead to splendidly life-like 
engravings of animals. Breuil also ascribes to the Perigordian period 
engravings in which the lines, at first shallow, gradually become deeper — 
as do those at Altamira, Pair-non-Pair and La Greze (Dordogne). These 
form a transition to the magnificent bas-reliefs in the Dordogne and 
Charente area. 
Solutrian-Magdalenian Up to the present day we know of no paintings that could unequivocally be 
ascribed to the Solutrian period. The early Magdalenian goes back to 
line drawings in black, which are frequently sketchy (Altamira, El Castillo, 
Le Portel and Niaux). To this group also belong the black tectiforms at 
Altamira. In the subsequent stage of development the line becomes wider 
and blurred; fine contour drawings executed with masterly skill are 
created; once again there is partial filling in, and the animals' coats are 
suggested by firm brush-strokes. This stage is best exemplified at Niaux, 
Pech-Merle and Le Portel. The body comes to be modelled more and 
more fully by application of colour, and is filled in with thin red and 
brown paint, or with black and red dots, as at Pech-Merle. The most 
advanced stage of Ice Age art, polychrome painting, has thus been reached. 
After fumbling beginnings it attains a magnificent peak of achievement, 
the first climax in the history of art, in the roof frescoes of Altamira, 
with their fine animal figures in shades of red and brown, outlined in 
black and accentuated by means of a burin. But Magdalenian art has 
now passed its zenith; it seems as though the Magdalenian artists had 
exhausted their powers. Ice Age art ends in an era of imitative work, 
of small line drawings which gradually become more and more stylized 
and schematized, until it eventually ceases completely in the Franco- 
Cantabrian area. 

Parallel with the development of painting one can observe an efflorescence 
of modelling in relief after the middle Solutrian and during the early 
Magdalenian. Magnificent examples are to be found at Roc de Sers 

24 



(Charente) and Cap Blanc (Dordogne). The twisted perspective has here 
been abandoned. Engravings executed with masterly skill are super- 
imposed upon older ones dating from the Perigordian. An animal's coat 
is suggested by fine hatching. The most splendid animal figures from this 
period are furnished by the 'sanctuaire' of the Trois Freres cave (Ariege). 
At Tuc dAudoubert (Ariege) and Montespan (Haute Garonne) the Mag- 
dalenian artists utilized the clay in the caves to create naturalistic works 
of animal sculpture. 

Before going on to discuss the implements and materials used in the 
execution of rock paintings a few words may be said with regard to the 
problem of lighting. Frequently the paintings are situated in dark lateral 
passages, often at an immense distance from the entrance to the cave, as 
is the case, for instance, at Niaux or Bedeilhac (Ariege). Often it is a 
hazardous undertaking to walk about in a cave — for example, in Tuc 
dAudoubert or Montespan. A light, as well as some sort of ladder or 
rope, are necessary to penetrate into certain caves. The problem of the 
lighting used by the artists cannot be answered simply. A few cupel- 
shaped lamps have been found, the best known of them at La Mouthe 
(Dordogne), which contained remains of some fatty substance. In the 
Pyrenean caves, particularly at Trois Freres, slabs of schistous sandstone 
have been discovered which are scorched at one end and coated with a 
carbonized mass. Presumably these slabs also contained a greasy substance, 
provided with a wick; this object must have been held in the hand by 
the opposite end and have served as a kind of torch. The use of torches 
is also suggested by the remains of charcoal that are to be found every- 
where in the caves. There is no doubt that the men who frequented the 
large caves were able both to provide lighting and to re-light a flame if it 
was put out by a draught or a drop of water. 

For the various branches of Ice Age mural art a variety of instruments and 
materials were required. Paintings were occasionally executed with dry 
crayon, but usually with a liquid paste. 

After the end of the Mousterian period, the last phase of Lower Palaeo- 
lithic culture, the strata often contain fragments of pigment. From these 
it appears that the Ice Age painter made most frequent use of ochre, from 
which he could obtain shades of colour ranging from a yellowish red 
through red to brown; in addition to this he used red chalk. Manganese 
ore and charcoal provided him with black for his palette. White 
was probably never used, while green and blue are entirely lacking 
in Ice Age painting. Violet tones have only been observed at Altamira. 
Red ochre has even been found sharpened into crayons, which we 



IMPLEMENTS 
AND TECHNIQUE 



25 



SIGNIFICANCE 
OF ICE AGE ART 



may assume were used in the manner of pastels. As a rule, however, the 
colours were ground to a fine powder to which oily or greasy substances 
were later added. Blood and albumen may have been used as binders. 
The paste thus made was then applied to the rock surface. Fingers, 
frayed twigs or tufts of feathers may have served in place of a brush. 
Occasionally the paint was also blown directly on to the wall with the 
mouth, a practice still found among Australian aborigines today. The 
distribution of fine grains of pigment over the surface of many paintings 
bears witness to the use of this technique. 

The engravings were cut into the rock by means of flint burins, such as 
are to be found in large numbers everywhere. In the case of some high 
reliefs which were cut deep into the stone stronger tools were required. 
Finds in early Magdalenian strata in the Dordogne point to the use of 
enormous stone picks. 
ORIGIN AND The origins of art in the Late Ice Age are shrouded in the deepest 
obscurity and the present state of our knowledge allows us to give only 
an outline sketch here. The development of art must undoubtedly have 
been preceded by the notion of the similarity between two living beings, 
whilst impulses must also have radiated from the oldest of the arts, that 
of drama. For already in his dances primitive man imitated living creatures, 
i.e., he copied their habits, gestures and mimicry. As is the case among 
primitive peoples still today, the mask or its image can become a complete- 
ly independent being conferring magic powers upon its wearer. 
Another impetus towards artistic creation may have come from the realm 
of hunting. Every day the Ice Age hunter came across the tracks left in 
the soil by wild animals, which could be imitated artificially. This may 
have induced man first to reproduce his own hands, and this may in turn 
have given him the idea of drawing lines on the clay with his fingers or 
painting on a rock wall with fingers steeped in clay. At first purely acciden- 
tal, such action will soon have acquired a purpose: a tangle of lines will 
gradually have led to arabesques and winding meanders, and at some 
point from these designs the silhouette of an animal will have been born. 
Homo sapiens The birth of art occurred in the Upper Palaeolithic age, some 30,000 years 
ago, when the last great glaciation spread over Europe. A new race of man, 
the so-called homo sapiens, immigrated into this part of the world, super- 
seding the older and more primitive human types. This man must have 
had latent within him the talent necessary in order to produce works 
of art. 

During the last Ice Age man lived amidst an environment dominated by 
mighty wild beasts, including mammoths, rhinoceroses, bisons, aurochs, 

26 




Polychrome standing bison, with hind quarters of three other bisons. Magdalenian. Detail from 
roof paintings in the picture gallery at Altamira. 



27 



wild horses, musk-oxen and reindeer, not to mention cave-bears, tigers 
and lions. Throughout his life man received strong visual and dynamic 
impressions from the dangers to which he was exposed in his daily 
struggle with these animals. From this also resulted his unique knowledge 
of the animal world, which alone was capable of inspiring artists to 
reproduce these creatures in monumental and lifelike form. 
The representation of animals on the walls of caves in France and Spain 
could, however, not have derived merely from the inspiration of a few 
individuals. This was no 'art for art's sake', although pure delight in 
beauty should not be excluded; but ultimately, as was the case also in 
later epochs, it existed against the background of the religious and social 
obligations and interests of the community to which the artist belonged. 
Ice Age art gives profound expression to the social and religious structure 
of the culture of the hunters who lived during that period. For the tribe 
to survive a certain stock of wild animals was always essential. There had 
to be sufficient quantities available, and it was a matter of concern that 
the stock of animals, decimated by hunting and natural causes, should 
continually be replenished. As in all hunter cultures, recourse was had 
to magic, to the performance of death and fertility rites. This is the only 
explanation for the fact that the paintings are located deep in the interior 
of caves and are often superposed on one another. For Ice Age man, as 
for hunter peoples of the present day or those that died out not long 
ago, the governing factor will have been similar: by confining the animal 
within the limits of a painting one subjected it to one's power in the 
hunting grounds. 

It may be assumed that the Ice Age artists, who were trained in a kind of 
'art school', were simultaneously the tribal magicians and the only men 
privileged to have access to the subterranean art stations. Here they per- 
formed the magic rites essential to the survival of the tribe. 

Six large We will now turn to the six caves at Altamira, Font-de-Gaume, Les 
Combarelles, Lascaux, Niaux and Trois Freres, which contain the finest 
works of Franco-Can tabrian art. Brief mention will be made of caves 
and rock-shelters in France, Spain and Italy. 
Altamira This cave is situated 19 miles from Santander, not far from the small 
town of Santillana del Mar in northern Spain. With its splendid frescoes 
it still takes pride of place, alongside Lascaux, among all the Franco- 
Cantabrian art stations — just as it did when it was first discovered almost 
a hundred years ago. Altamira extends for 300 yards into the heart of 
a limestone massif, but the famous paintings are to be found in the so- 

28 



Fig. I — Hind. Engraving. Al- 
tamira. Length 7 ft. 4Y2 in. 




called 'picture gallery' only about thirty yards from the entrance. Its 
height is in some parts no more than six or seven feet, so that one can 
best view the paintings when lying down. This explains why the child 
Maria Sautuola was the first to notice them. In the deep interior of 
the cave there are other paintings in black and red, as well as engrav- 
ings originating from different periods, distributed irregularly over the 
walls. 

The frescoes on the roof of the picture gallery are better preserved than 
any of the others. There are 25 polychrome figures here, most of them plate p. 27 
painted in red ochre, and the others in brown and black. The animals 
are painted almost life-size and measure fifteen yards in length. They 
are mainly bisons, their contours being accentuated in parts by engrav- 
ing. Between them other animals can be identified with slightly modelled 
body surfaces and contour lines in black, superimposed on even older 
pictures. The roof is covered by numerous signs, partly club-shaped and 
partly scalariform. On the right-hand side of the frieze there are to be 
seen several unfinished polychrome paintings, as well as older animal 
figures in a light shade of red, groups of dots and a small series of painted 
human hands. 

The technique of these polychrome paintings attains a higher degree 
of perfection than any others in Franco-Cantabrian art. For painting 
the bodies of the animals the artists made use of charcoal and ochre, 
which afforded them shades of colour ranging from yellow through red 
to brown. Remains of pigment sharpened like crayons have been found 
in the archaeological strata of the cave. The outline of the pictures was 
first of all drawn with a fine black line. In this way the artist obtained a 



89 



Plate p. 31 



rough sketch which he could now proceed to fill in with colour and model. 
Shortly before the work was completed certain details such as the eye, 
horns, nostrils and parts of the hooves were accentuated with a burin, 
which in most cases was also used beforehand to make the contour line 
wider. After the colour had been applied, the paintings must sometimes 
have been washed and scraped to produce finer shadings of colour and 
still more delicate and harmonious effects. With such simple means the 
Ice Age artist was able to model the surface of the body, and to distrib- 
ute light and shade by delicately contrasting the colours. He succeeded 
in depicting the animals in their habitual attitudes: here we see almost 
life-size bisons, some standing erect, while others are portrayed lying on 
the ground, galloping at full speed, or slowly stealing along. But the 
artists did not rest content with this. There is another factor that contrib- 
utes to the moving vigour of these animal figures. The roof of the large 
chamber is not flat and smooth, but has bulges, small protuberances and 
uneven patches. With the aid of paint the artist was able to incorporate 
the natural relief of the rock into the bodies of the animals in such parts 
as the head, rump or crupper, thus making them extremely realistic. 
At Altamira the representations are mostly of bisons, followed by wild 
horses, hinds, stags, ibexes and wild boars, less frequently by aurochs, and 
extremely rarely by elks and wolves. 

All these animals are painted in a naturalistic style, and with elaborate 
care and accuracy. The bisons range in length from 4 ft. 6 in. to 6 ft. 
The magnificent hind to the left of the series of frescoes is 7 ft. 4 in. in 
length. 

One would look in vain for scenic compositions at Altamira. These are 
almost completely absent from Franco-Cantabrian art. Each animal por- 
trayed on the roof of the gallery forms an entity of its own; although 
they are sometimes depicted in groups side by side, there is no scenic 
connection between the individual animals. 

Superpositions are rarely to be found in the gallery. The polychrome 
figures constitute the most recent layer in the Altamira paintings. In 




Fig. 2 — Crouching bison. Paint- 
ing. Altamira. Length 5 ft. 1 in. 



3° 




Head and neck of a polychrome hind. Below, on the right, small bison in outline. From the picture 
gallery, Altamira. Length of hind 7 ft. 4I/2 •'»• 



31 



many places they have been painted on top of older figures and are not 
themselves covered by paintings or engravings. At the most some of the 
comb-shaped red signs between the polychrome paintings may be thought 
to date from the same period, or to be more recent. 
The polychrome paintings are the work of the Upper Magdalenian 
period, which may be dated circa 12,000 B.C. 

Altamira also harbours some extremely fine engravings; one may merely 
mention the magnificent heads of hinds, bisons and stags, all of them 
belonging to the later Magdalenian period. 
Fontde-Gaume Font-de-Gaume, one of the most lavishly decorated art stations on French 
soil, is situated in the Beune valley, a side-valley of the Vezere, approx- 
imately half a mile from Les Eyzies. A narrow corridor measuring over 
100 yards in length, it extends deep into the mountain and has several 
lateral galleries branching off from the main passage. The main gallery 
in parts reaches the considerable height of 23 to 26 ft. The first pictures 
are found 70 yards from the entrance at a spot where daylight no longer 
penetrates. Some works nearer the entrance may possibly have been 
destroyed in the course of time by atmospheric influences. At Font-de- 
Gaume some 200 pictures can be deciphered, though at times only with 
difficulty. Among the best representations is one of the finest friezes of 
the cave: a series of polychrome bisons, on some of which small, very 
delicately engraved mammoths are superimposed. These animal pictures 
are accompanied by tectiforms, which are unfortunately in a poor state 
of preservation. The bisons are painted in red and brown, the horns, 
eye, dorsal line and muzzle being accentuated by means of deeply engraved 
strokes. 16 feet further along one may discern parts of a fine frieze con- 
tinuing along the left-hand wall as far as the 'chamber of small bisons'. At 
one point there is a rendering of two magnificent reindeer facing one 
another. From their antlers and their demeanour they appear to be male 
and female: the buck seems to be scenting the head of the hind. In this case, 
too, use was made of red and brown to fill in the bodies and of the burin 
to accentuate the contour lines. Another frieze of polychrome bisons 
adorns the right-hand wall of the main gallery. This also comprises fine 
reproductions of several small horses, a wolf and a reindeer. In the 
'chamber of small bisons' the walls and roof are covered with pictures of 
these animals. Some of them are painted in a uniform black shade, or in 
brown, and there are also figures painted in more than one colour. The 
remains of oxen painted in black and brown seem to date from an earlier 
epoch. After this chamber the main gallery comes to an end in a narrow 
corridor, the walls of which are covered with a variety of polychrome 

32 



pictures. Noteworthy among them is the engraved figure of a feline animal 
facing several horses, and particularly that of .a rhinoceros, a typical 
denizen of'Europe during the Ice Age, reproductions of which are rarely 
met with in cave art. The woolly-haired rhinoceros is outlined in red, 
the hair being suggested by hatching along the contour line. The horn 
is clearly visible. This figure is undoubtedly from the Aurignacian period. 
On the right-hand wall of the main gallery there is also a picture of a 
black reindeer that is worth mentioning, although its head can hardly Plate p. 34 
be identified; to the right of it there follows a bison, likewise painted in 
black, the hind quarters of which are formed by the relief of the rock. 
In Font-de-Gaume tectiform signs are repeatedly found in various shapes, 
some painted and others engraved. According to Breuil these may be 
renderings of Palaeolithic dwellings, the roofs of which were covered with 
branches or dried grass. 

The works of art at Font-de-Gaume derive from several prehistoric periods 
during which changes took place not only in the human settlement but 
also in the fauna and flora of the area. There are numerous superposed 
pictures at Font-de-Gaume, making it possible to distinguish clearly the 
chronological sequence of the various styles of painting. This shows that 
the cave was frequented by human beings during all periods of the Upper 
Palaeolithic. Breuil's summary of the animal types depicted at Font-de- 
Gaume is of interest: according to him there are 80 bisons, 40 wild horses, 
23 mammoths, 17 reindeer and cervidae, 8 aurochs, 2 rhinoceroses, one 
or two felines, one wolf and one bear. Breuil believes that rhinoceroses, 
felines, bears and ibexes inhabited the Dordogne area during the first 
third or quarter of the Upper Palaeolithic; the aurochs was extant from 
the beginning but not permanently; the reindeer was likewise extant 
during all periods, whereas the bison was only seldom featured at the 
beginning, but predominates over the other animals towards the end. 
The rhinoceros, feline, cave-bear and possibly also the ibex became extinct 
before the end of the Ice Age or else migrated to other areas; the mammoth 
reappeared from time to time after periods of absence, and horses were 
particularly plentiful at the beginning of Ice Age art. Breuil dates the 
pictures to the early and middle Magdalenian period, while some of them 
may already belong to the Aurignacian-Perigordian cycle. 

The cave of Les Combarelles is situated in a precipitous limestone rock Les Combarelles 
face not far from Font-de-Gaume, only a few miles from Les Eyzies. The 
cave consists of two low narrow galleries, which meet in a fairly spacious 
antechamber. It is, however, only the left-hand gallery that has yielded 
mural art of importance. The cave extends for over 250 yards into the 



33 



heart of the mountain; at one time it was a subterranean stream, which 
had, however, dried up by the time of Ice Age man. Les Combarelles is 
one of the few caves where finds are restricted to engravings, the traces 
of painting being quite insignificant. The first of the pictures, which 
number several hundreds, adorn the walls some 75 yards from the entrance. 
It is difficult to pick out the individual figures from the vast tangle of 
lines and strokes. Out of 300 pictures 291 have been deciphered, while 
some 100 representations have only been preserved in part and can no 
longer be identified. Breuil puts the number of figures counted at Com- 
barelles at 116 horses, 37 bisons, 19 bears, 14 reindeer, 13 mammoths, 9 
ibexes, 7 head of cattle, 5 stags, 3 hinds, 5 lions, 4 wolves, one fox and 
39 figures of human beings. 

When one examines the pictures at Combarelles in greater detail the 
contours and lines suddenly come to life, and there appear magnificent 
animal figures that rank among the finest products of the Ice Age. 



Black bison. On the left a reindeer walking away. Magdalenian. From the main gallery, Font- 

de-Gaume. 




34 



Drawn in a flowing style, the majority of the pictures probably belong 
to the Magdalenian period, but at the same time-one should not overlook 
the more ancient and rigid pictures dating from the Aurignacian. 
Some of the engravings of mammoths and horses are magnificent. In the 
case of the horses alone Breuil distinguishes between four different species. 
An enormous bear conveys an impression of great strength. The splen- 
did feline, now famous, can easily be identified, despite a thick covering 
of calcareous deposit. Its hind quarters are raised up; the upper part of 
the thigh is vigorous and muscular, the shoulder-blades project outwards, 
and the body is squat and strong. The head, as well as the chest, paws 
and stomach are all worked in bas-relief. There are some other pictures 
of felines at Combarelles, but an exact identification of these animals 
is usually very difficult, since they really resemble lions rather than tigers 
and are rarely portrayed in Franco-Cantabrian art. Mention may also be 
made of a striding reindeer and a mammoth with thick fur and an involute 
trunk. 

The human beings depicted at Combarelles, though no more than medio- 
cre from an artistic point of view, are all the more significant from the 
standpoint of cultural history. As is the case in other caves, they give the 
impression of men wearing animal masks. Among the most grotesque 
figures is a human silhouette with a mammoth's head and greatly elongated 
arms, which may perhaps suggest tusks. Elsewhere a male figure appears 
to be following a woman, and the head of a bearded man is distinctly 
discernible. There must be some deeper reason for the fact that these 
figures of human beings are always represented in an unskilful manner, 
whereas elaborate care is bestowed upon the pictures 
of animals. One of the reasons may be the great 
reluctance of primitive man to depict his own image 
and thus place himself in the hands of a magic- 
working enemy. In the majority of cases, however, 
the idea was to represent masks used in hunting or 
rites which personified mythical beings and were 
connected with some magic cult. Scenes of a phallic 



Fig. 4 — Lion. Wall engraving. Les 
Combarelles. Length 2 ft. 3I/2 in. 




35 



tf**% 




Fig. 3 — Horse. Wall engraving. Les 
Combarelles. Length 2 ft. pl/ 2 •»*■ 



nature are rare but, where they occur, according to 
Breuil, they may bear relation to magic fertility rites. 
All the finest engravings at Combarelles are assigned 
by Breuil to the early and middle Magdalenian 
period. Combarelles serves as a perfect illustration of 
the fact that the cave could not possibly have been used 
for human habitation. Leaving the entrance out of account, the narrow 
interior of the cave is dark and damp, and settlement there can never have 
been feasible. The pictures on the walls will therefore not have been drawn 
for decorative effect or purely for their own sake. Only a few members of 
the tribe could have been present in the narrow corridor at the same time. 
Combarelles shows more clearly than many other caves that the pictures 
must have been produced as part of some magic ritual performed by a 
few select representatives of the tribe. 
Lascaux Lascaux was not discovered until the relatively recent date of 1940. For 
this reason it appears to surpass everything previously attained in Franco- 
Cantabrian mural art. The pictures are in a wonderful state of preserva- 
tion, and the dazzling brilliance of the colours is hardly credible to an 
observer coming from caves in which the paintings have already faded. 
The pictures in the vast chamber and the adjoining gallery are applied 
upon a light shimmering ground which sets off in strong contrast the 
various shades of red, yellow, brown and black. 

The cave is situated in a limestone massif above the Vezere valley, just 
over a mile from the village of Montignac. Directly upon entering we 
find the picture gallery, which measures 33 yards in length and 11 yards 
in width; its walls are adorned right up to the roof with magnificent animal 
paintings. At the end of the chamber a small gallery opens up, which 
continues in the same direction as the chamber and loses itself in the 
inner depths of the mountain; its walls and roof are also decorated with 
magnificent frescoes. On returning to the main chamber one's eye alights 
upon the opening to a gallery in the right-hand wall, leading to a part 
of the cave situated at a somewhat higher level: this part of the cave has 
numerous engravings. Right at the rear the 'apse' of the so-called 'nave', 



36 




Two small horses, with hind legs of a third animal. Above, on the left, part of a black cow, 
painted over an older figure in red, is just recognizable. Aurignacian-Perigordian. Lascaux. 
Width 5 ft. 6 in. 



an extension of the gallery in the shape of a chamber, leads to a shaft 
some 23 feet deep through which the visitor clambers down into a lower 
gallery. In this part one of the few narrative compositions of Ice Age art 
has been discovered: a badly wounded bison, its flank pierced by a lance, 
its horns lowered ready for the charge, and before it, sketched as always 
with a few strokes, the figure of a man is seen prostrate on the ground; 
in the foreground is a bird on a perch, and to the left a rhinoceros can 
be seen moving away. The whole composition is drawn in black, with 
the lines slightly blurred. It has been interpreted in many different ways; 
it is probably either a tragedy of the chase or a scene of a magic ritual 
character. 

Of the paintings in the picture gallery and the adjoining gallery the 
finest examples may be mentioned here. The large chamber is also referred 



37 




Fig. 5 — Galloping horse pierced by 
arrows. Lascaux. Length 4 ft. 7 in. 



to as 'the chamber of the aurochs' after the very striking figures portrayed 
here. Three gigantic pictures of such aurochs and parts of a fourth 
animal form the decoration around the walls. These figures, measuring 
up to 18 ft. in length, are unique phenomena in Ice Age art. The contours 
are painted in black, with flowing lines; the inner surface of the bodies 
is filled in with black colour, or black dots, particularly along the belly- 
line, the muzzle and legs. The horns and hooves of the aurochs are twisted 
round and treated in the same plane as the body, i.e., the painter could 
not as yet visualize them in normal perspective, but represented them in 
twisted perspective, this being a characteristic feature of the Aurignacian- 
Perigordian style. These aurochs overlap with an older group of wild 
oxen, painted over evenly in dark red. Stylistically the so-called 'unicorn', 
the most curious animal at Lascaux, must be ascribed to this group of 
aurochs. It is the first figure on the left-hand wall of the large chamber. 
The contour lines are rugged and sharp, the body and legs are powerful, 
and the animal gives the impression of being with young. It bears most 
resemblance to an ox or a rhinoceros, but the small head, from which 
two long straight shafts protrude, belongs wholly to the realm of legend. 
The chest and back are dappled and the tail is quite short. This is either 
some mythical animal or a human being in animal disguise. It is not the 




vS. 



Fig. 6 — 'Swimming stags'. Las- 
caux. Length of frieze 16 ft. 4 in. 



Fig. 7 — Small stags. Antlers in twisted perspective. 
Lascaux. Average length of animals 2 ft. 9I/2 in. 




only mythical creature to be found in Franco-Cantabrian art, but probably 
the most striking one. 

Between the two aurochs that face each other on the left-hand wall 
there are several small stags with fully-developed antlers; they too are 
painted in dark red, and seem to be contemporaneous with the small oxen. 
Superposed on the aurochs nearest to the unicorn is a fairly large horse, 
with the body painted in a dark reddish brown, but the head, mane and 
legs in black. In front of it, and opposite the head of the second ox, are 
the head and back of a similar horse. Below these horses we can see a 
group of relatively small, very dark galloping horses, which resemble the 
black ponies in the adjoining gallery. Here the shaggy horses are topped 
by a large cow that appears to be leaping towards a lattice-like sign. It 
is painted in black over an older figure in red. The superimposition of 
the two colours gives the animal an almost polychrome character. Also 
worthy of note in this gallery are the very fine horses with small dark 
heads and bodies painted in light red, on some of which the hair is in- 
dicated, as well as several cows in reddish brown. In between them are 
remains of older figures as well as a great number of these lattice-like 
signs in several variations. Were they supposed to suggest traps or game 
preserves? 



Plate p. 3 



Plate p. 37 



Fig. 8 — Large cow. Painting in red 
and black. Lascaux. Length ft. 2I/4 in. 




39 



In the lateral gallery engravings prevail, although paintings are not 
Fig. 6 infrequent. The picture of several stags swimming has become famous: 
the animals appear to be crossing a river one behind the other; only the 
contours of the heads, in black, with gigantic antlers, and parts of the 
necks are shown. (The herd measures 16 ft. 6 in. in width). On an opposite 
wall there are several wild horses and two fine bisons in dark brown, the 
flank of the animal on the left being in red; the animals stand with their 
hind legs opposite one another. 

Prominent among the engravings in this part of the cave is a mighty stag. 
Also noteworthy are an elaborately engraved head of a horse and a very 
fine lion. The outlines of the black horses and polychrome lattice-like 
signs are engraved. 

At Lascaux, in addition to the well-known painting techniques in which 
the artist employs his finger or a brush, finely pulverized pigments were 

Two black bisons with black and red arrows drawn on their bodies. Above, on the left, a third 
bison, of which the dorsal line is formed by a rock ledge. Below, on the left, a large horse, and 
in the centre a small equine animal. Salon noir, Niaux. Length of bisons 3 ft. 11 in, and 4 ft. 2 in. 




40 



blown on to the rock. If one examines the contours of many animals exact- 
ly, one can see that these consist of round spots which merge into one 
another. This technique, hitherto only known from silhouettes of hands, 
was chiefly employed for the colouring of horses' manes. The 'twisted 
perspective', a feature absent from Magdalenian and Upper Solutrian 
art, shows that the Lascaux pictures can be assigned to the later Aurigna- 
cian-Perigordian period, an attribution also borne out by other stylistic 
features. The art of this epoch reaches its zenith at Lascaux, and attests — 
apart from some paintings, where primitive features are still evident — 
to a masterly skill in the execution of the pictures, which occasionally 
attain monumental proportions. The variety of the techniques that follow 
upon one another at short intervals suggests that these experienced and 
imaginative artists must have been possessed by a kind of creative frenzy. 
The art of Lascaux is indeed great, and in its way perfect. 
This gigantic cave, overlooking a valley in the Pyrenees, is situated 2y 2 
miles from the small town of Tarascon-sur-Ariege. Tourists and local 
inhabitants have flocked to visit it since the 17th century, and have left 
their names upon its walls. The paintings in the famous 'salon noir' were 
discovered by Dr. Garrigou as early as 1866, but no special significance 
was attached to them. It was not until 1906, after Ice Age art had been 
recognized as authentic, that attention was once again paid to these black 
line paintings, which are in an excellent state of preservation. 
Passing along a narrow corridor, the visitor to Niaux arrives at the first 
chamber, which has a small lake, usually filled with water. 668 yards from 
the entrance a corridor leads into a vast chamber, the roof of which 
gradually becomes lower. In this part of the cave one may already find 
a number of varied signs, such as red and black dots or groups of lines. 
The visitor proceeds through chambers of red and yellow marble, which 
is often brightly polished; elsewhere stalactites may be seen cascading 
down from precipitous walls and roofs that lose themselves in the darkness. 
After a chamber filled with various sorts of sand, the visitor finds himself 
in the 'salon noir', which has the finest pictures in the cave. It was at this 



Plate p. 40 



Kg. 9 — Bison with arrows drawn 
on. Niaux. Length j ft. 3 in. 




4 1 




Head of bison, engraved in clay. Magdalenian. Niaux. Total length of animal i ft. 10% in. 



42 



spot that one of the explorers, Molard, came across flint implements and 
remains of bones in a state of decomposition. All the pictures are drawn in 
a wonderful flowing style with black contour lines; the hair of the animals 
is indicated by hatching, and the horns and hooves are rendered in 
natural perspective. These pictures, of the bisons in particular, as well 
as those of the horses, ibexes and stags, are among the finest examples 
of Magdalenian art. Breuil assigns them to the middle and late Magda- 
lenian periods. The figures of an earlier date were drawn in the linear 
technique. Later, as the style became more fluent, artists also made use 
of rock projections to model animal bodies in relief. The most important 
group of pictures in the 'salon noir' is that of several bisons, which appear 
to have been pierced by arrows — apparently connected with a magic 
hunting rite. A rare feature in Franco-Cantabrian art are the engravings 
at Niaux, cut into the clay. Thus on the floor of the 'salon noir' the 
discovery was made of engravings of a particularly fine bison as well 
as two trout, one of the best renderings of fish handed down to us from 
the Ice Age. At the end of the 'salon noir', which is situated 840 yards 
from the entrance, the cave continues into the innermost recesses of the 
rock, where one still finds some remains of bison figures in black, as well 
as a few signs and animal pictures in red. Right at the end of the cave 
of Niaux there is a placid lake; not the slightest movement of air ripples 
its calm surface. 

This cave is situated near Montesquieu-Avantes, on the estate of Count 
Bdgouen. In 1914, two years after the discovery of the cave of Tuc 
d'Audoubert (see below), Henri Begouen and his three sons discovered 
the second of the two great art stations on their land: the cave of the Three 
Brothers, or Trois Freres. Deep in its interior, in a chamber known as 
the 'sanctuaire', are the most important paintings of the cave. This cham- 
ber, the floor of which drops steeply, has on its walls a vast number of 
overlapping engravings, partly originating from the Aurignacian arid 
Perigordian, and partly from the best phases of the Magdalenian period. 



Plate p. 42 



Trois Freres 



Fig. 10 — Ibex. Niaux. Length 1 ft. 81/ 2 in. 




43 




Fig. j i — Trout. Engrav- 
ing in clay of cave floor. 
Niaux. Length 111/4 in. 



The lines of the engravings are often cut deep into the rock, so that the 
dark incisions stand out like cameos against the light background. Here 
Breuil has made copies of some magnificent bison figures, mighty stags, 
a large number of reindeer, bears' heads, horses, ibexes, and at the entrance 
to the 'sanctuaire' two large lions' heads, viewed from the front, which 
suggest two guardian figures. On the right-hand wall of the chamber lies 
a mammoth drawn in an archaic manner, with the dorsal line breaking 
off sharply, and a bear pierced by many holes, with blood pouring forth 
from its muzzle. In other parts of the cave explorers came across a rendering 
of two snowy owls from the Aurignacian period. 

In one part of the 'sanctuaire' one can distinguish a bison with human 
hind quarters and behind it a dancing figure wearing a bison's head, but 
depicted in an erect attitude. This is presumably a human being in animal 
disguise, who is shown holding a longish instrument in his hands and 
putting it to his mouth. This may possibly be a flute. There are several 
'magician' figures of this type in the 'sanctuaire'; between the animal 
paintings there are also small distorted human faces, which Breuil inter- 
prets as animals' souls. It is here, too, that we find the most curious ren- 
dering of man-and-beast in Ice Ace art: the 'magician of Trois Freres', 
who looks down from a height of 13 feet on one wall of the 'sanctuaire', 
seemingly ruling over the whole vast concourse of animals and half- 
human, half-animal creatures. 

One stag has a long beard hanging right down to its chest and small round 
eyes staring from its face; its forelegs are raised up and its hind legs 
perform a dance; the genital organ is clearly marked and the head is 
crowned by gigantic antlers. Black brush-strokes serve to accentuate the 
individual parts of the body, but the contours, on the other hand, 
are engraved. The Magdalenian artist may have regarded this figure 
as the focus of all his paintings. Was it the Great Spirit, master of 
the animals, who disposed of all matters pertaining to hunting and 
fertility? 



44 



The vast cave of Baume-Latrone is situated some 814 miles from Nimes 
on the left bank of the river Gard. Its paintings were discovered in 1940. 
All of them- are housed in a chamber situated 260 yards from the entrance. 
First of all a number of representations of human hands, mainly left hands, 
were noticed on the roof of this chamber. Soon explorers also observed 
the curious paintings. They are animal figures, quite archaic in style, 
drawn on the wall with fingers dipped in clay, recalling those found 
in the Andalusian cave of La Pileta. We can distinguish some six or seven 
elephants measuring up to 4 ft. 9 in. in length, as well as a rhinoceros 
and a snake. The trunks of the elephants are rendered by strange zigzag 
lines; the snake is 9 ft. 9 in. long, and its head resembles that of a bear 
with its jaws opened in a threatening manner. At Baume-Latrone there 
are also some representations dating from a more advanced phase of 
development, executed in a linear style, but both they and the paintings 
done with the fingers may probably be ascribed to an early stage of the 
Aurignacian. 



CAVES AND 
ROCK-SHELTERS 
IN FRANCE 

Baume-Latrone 



Plate p. 45 



1 




v-; ..-.-._ 







Elephant, drawn with several fingers. Early Aurignacian. Baume-Latrone. 



45 



Chabot In 1878 the teacher L. Chiron discovered the cave of Chabot, situated near 
the town of St. Martin d'Ardeche. In an antechamber where daylight 
could still penetrate the explorer came across numerous deeply-incised 
engravings. But it was not until Ice Age art had found recognition that 
they were brought to notice and deciphered. They are engravings of 
mammoths; in 1928 Breuil discovered more figures of horses, ibexes and 
mammoths, all of them archaic in style. Similar engravings are to be found 
in the adjoining cave of Le Figuier. Here, as at Chabot, Upper Palaeolithic 
strata were encountered. 
Ebbou The pictures at this cave, also situated in the Ardeche gorge, were discov- 
ered by Abbe Glory in 1946. In addition to the representation of a hand 
in red near the entrance, Glory came across 70 engravings in a chamber 
71 yards long and about 16 yards wide; he was able to identify 24 horses, 
12 aurochs, 2 bisons, one mammoth and several ibexes. Each of these 
figures has only one pair of legs; the horns of the aurochs are depicted 
in twisted perspective, and a sense of perspective is also lacking in the 
treatment of the stags' antlers. 

While dealing with Ebbou mention may also be made of engravings in 
a similar style found in the Ardeche caves of Colombier and Oullins. Not 
far from the Pont-du-Gard, approximately 6 miles north-east of Nimes, 
Abbe Bayol found in a cave which has been named after him a line 
drawing of an ibex in red as well as several representations of hands. These 
paintings and other remains of red pigment point to an archaic style. 
Le Portel This cave is situated close to the farm of that name, not far from the 
railway-station of Varilhes. The paintings were discovered in 1908 by 
Dr. Jeannel. The owner of the cave, M. Vezian, carried out several excava- 
tions in the course of which he recovered finds dating from the middle 
Magdalenian period. It is not easy to gain access to the cave; it is entered 
through a narrow low corridor which descends sharply; the floor is damp 
and clayey, and it is only at the back of the cave that it becomes more 
even and easier to walk upon. It is here that the paintings are to be found; 
they are in a corridor 65 yards long leading into some vast chambers, 
from which several galleries branch off. At the rear of the first corridor 
on the left-hand wall there are various niches in which several signs in 
red have been observed, one of which could be the representation of a 
large hand. Next to it is a recumbent red reindeer, painted in a linear style, 
with the antlers shown in 'twisted perspective'. The right-hand wall has 
several almost faded fragments of pictures, the head of a bison outlined 
in black, an owl with a large round head and disproportionate body, and 
a fine black pony. 



46 




Two bisons. On the left, hind quarters of a third animal. Magdalenian. Breuil gallery, Le Portel. 



In the following left-hand gallery there are two pictures of human beings 
with grotesque faces. There are also some fine horses painted in a soft 
brown colour; their archaic style recalls the dappled horses of Pech-Merle 
as well as some figures at Lascaux. The central gallery, named after 
F. Regnault, contains a small horse painted in red and a badly drawn ox. 
Almost all the pictures in this part of the cave are painted either in black 
or sepia, and judging from their style may be dated to the Aurignacian- 
Perigordian period. Figures of horses predominate, but bisons are no 
rarity. The finest representations of horses at Le Portel are located at the 
rear of this gallery; they are painted in black, with the paint encroaching 
upon the inner surface of the animals' bodies, giving the effect of mod- 
elling; this makes it possible to attribute it to the middle Magdalenian. 
Further along is the Breuil gallery, named after its discoverer, where 
numerous scratches testify to the fact that it was once inhabited by cave- 
bears. On the right-hand wall in the front part of the gallery there are 



47 



several engravings executed in the Magdalenian style, among them a fine 
rendering of a bison and one of a horse pierced by an arrow. The picture 
of two bisons facing each other may be accounted one of the finest works 
at Le Portel. They are outlined in black, and the body of the animal on 
the right shows a tendency towards modelling through the spreading 
of the paint inwards. The horns are drawn in natural perspective. Breuil 
dates these animals to the early Magdalenian. 

Thus the two great art cycles are clearly represented in this cave: relatively 
early Aurignacian-Perigordian and early Magdalenian. 
Tuc d'Audoubert On the estate of the Begouen family near Montesquieu-Avantes there is, 
as well as Trois Freres, the vast cave of Tuc dAudoubert. The three 
Begouen brothers penetrated into the cave, which is watered by a small 
stream, the Volp, by means of a home-made canoe. The Volp flows into 
the open out of one chamber of the cave after having cut its way for just 
over a mile through the subterranean rock, and it was thus that the cave 
was formed many millennia ago, before it was frequented either by cave- 
bears or by man. It was from this chamber that the explorers succeeded 
in penetrating into the cave. In one small gallery they came across fine 
engravings of horses and bisons, a small reindeer, and several arrows and 
club-shaped signs. But the most significant works were found deep in the 
interior of the upper cave: in the centre of the last chamber, 765 yards from 
Plate p. 62 the entrance, were two unique sculptures of bisons, propped against a 
ledge of rock. They are male and female: the bull is depicted following 
the scent of the cow. Both sculptures, fashioned by the artist in the soft 
clay of the rock floor, have something fascinatingly lifelike about them 
and thousands of years ago must undoubtedly have served a purpose in 
some fertility rites performed here in the interior of the cave. There is 
another characteristic which marks this chamber out as a place of worship 
during the Ice Age: in the clay floor marks were found left by human feet, 
which judging by their size must have been those of youths no more than 1 5 
years of age. These footprints, the surface of which was covered and protect- 
ed by a hard coating of clay, automatically bring to mind the initiation 
ceremonies that play so important a part in tribal life, when the youth who 
has attained manhood is accepted into the adult community. Visitors to the 
cave, which incidentally is not greatly frequented owing to the difficulty 
of access to it, report that they experienced an eerie and stirring feeling 
when they entered this ancient place of worship of Magdalenian man. 
Montespan The cave complex of Montespan extends for over half a mile as the crow 
flies between the villages of Ganties and Montespan. The galleries 
measure in all some 2750 yards in length. The cave is watered by a small 



48 



river, the Hountao, which makes movement within it extremely difficult. 
The engravings and sculptures were discovered in 1923 by the explorers 
N. Casteret-and H. Godin. Parts of the cave are today permanently under 
water, which was probably not the case during the drier Magdalenian era. 
The upper part of the cave, in the direction of Ganties, yields the first 
engraving after 230 yards. This is a rearing horse; there are also figures 
of three horses, a mule and a bird. The horns of the 8 bisons are curved 
and rendered in natural perspective. In another dry gallery several inter- 
esting engravings can be noted, e.g.: a fine horse's head with a mane elab- 
orately engraved in clay and a delicately curved neck-line. Further along 
on the right-hand side the wall gives the impression of being perforated by 
spearheads, and other similar holes, cut rather crudely deep into the clay, 
pierce the horse depicted close by. Undoubtedly we have here traces of 
some magic practice: weapons must have been thrown at these animal 
pictures to ensure the success of a hunting expedition. 




Back view of a headless bear, modelled in clay. Montespan. Length 3 ft. y 1 ^ in. 

49 



The second part of the cave, situated at a lower level, is of greater signif- 
icance from an artistic point of view but is only accessible if the visitor 
is inclined to take a bath in the icy water of the river. After covering a 
distance of some no yards he finds himself in a gallery which is no longer 
under water. On the walls of this gallery, which measures 175 yards in 
length, numerous engravings are distributed irregularly. Owing to the 
damp they are not in a good state of preservation. Breuil mentions 4 
complete horses, 4 bisons and one bovine animal. In addition to these 
he deciphered several heads of bisons and horses. All the engravings at 
Montespan date from a relatively early Magdalenian period. 
But the highlight of this cave, as at Tuc dAudoubert, are the clay sculp- 
tures. There are several less significant specimens, such as the horses, 
that should rather be classified as bas-reliefs; there are some sculptures 
which have been called lion statues; others have deteriorated so badly 
that they have completely lost their shape, and now only a heap of clay 
remains to testify to their past existence. But one small, relatively low 
chamber contains a sculpture modelled in the round representing a bear, 
of which the head, however, is missing. This sculpture measures 2 ft. in 
Plate p. 49 height and 4 ft. 8 in. in length. The animal is rendered squatting, with 
its front paws stretched out and its hind legs pulled up under its belly. 
It is covered with a thin layer of sinter deposit. In the middle of its neck 
is a large hole, probably made for a wooden plug with which a real 
bear's head must have been affixed on top instead of one also modelled 
in clay. When the wood rotted the bear's head fell to the ground; the 
remains of this were found below the sculpture by the explorer Casteret. 
Once again the heavy perforation of the body is striking. This may 
probably be attributed to spears having been hurled at it as part of a 
magic rite. 

From an artistic point of view this statue is of less value than those 
of Tuc d'Audoubert, but it is unique for the light it throws upon the 
magic rites practised by the Magdalenian hunters. It supplied Breuil 
with a solution to the problem why the figure of the bear at Trois Freres 
is studded with holes. 
Gargas This cave is situated on the parish boundary between Aventignan and 
St. Bertrand de Comminges. The hill into which the cave extends lies 
above the left bank of the Garonne. Gargas is a great tunnel with several 
chambers. It was used at different times by man as a shelter. Since the 
19th century it has frequently been visited by tourists. From 1887 onwards 
F. Regnault made excavations in the cave and within a short space of 
time he found numerous skeletons of cave-bears and other animals that 



50 




1 1 H 




. 


■f £ ! 




■' r . #& •■ 








Red and black silhouettes of hands. Early Aurignacian. Gargas. 



had met their end there. The excavations revealed a constant succession 
of Aurignacian and Perigordian strata. 

A particular feature of this art station is the wealth of silhouettes of 
human hands to be found there. Shortly after entering the cave, in the 
first chamber, one's eye is caught by the silhouettes of human hands in 
black and red scattered over the wall on the left. Some of them, especially 
those near the entrance to the cave, are much faded, but in the interior, 
on the other hand, they are in a good state of preservation. There are 
also a great number of representations of hands on the walls of the three 
adjoining chambers. There are some 18 art stations in the Franco- 
Cantabrian area in which it has been possible to identify this curious 
element of Ice Age art. But nowhere are they so numerous as here at 



Plate p. 51 



Gargas. The hands painted in red, black and yellow are the only coloured 
representations to be found in this cave, which is almost permanently 
damp. Two types of representations of hands may be distinguished in Ice 
Age art: one of these is the negative, or silhouette of the hand, such as 
is to be found at Gargas. This type is produced by the artist pressing his 
hand against the wall and blowing pigment from his mouth, or through 
a tube, on to the surrounding surface. In this way a sort of shadow-figure 
or silhouette is formed, i.e. the light colour of the hand stands out clearly 
against the brightly-coloured background, the fine-grained texture of 
which can only have been produced by blowing. Much less frequent, by 
contrast, are positives of hands. These were produced by the artist dipping 
his hand in paint and stamping it on the wall. Curiously enough, this very 
ancient custom of depicting hands is not restricted to Franco-Cantabrian 
Ice Age art, but is also found in numerous hunter cultures in America, 
Africa and Australia. It presumably served primitive man as a sort of 
personal symbol, a kind of signature which defined his relationship with 
other members of the tribe or with supernatural forces. 
More than 150 representations of hands have been identified at Gargas. 
Those in red are frequently superposed by black ones, and are inferior 
in execution; they certainly date from an earlier period. In the case of 
hand silhouettes, it is generally the left hand that is shown, whereas in 
the case of positives it is generally the right hand. One explanation for 
this could be that, when making a negative, the pigment or the container 
for it was held in the right hand — assuming that Ice Age man was right- 
handed as we are today — and thus only the left hand was free for applic- 
ation, whereas to make a simple impression the free right hand could 
easily be used. 

A unique feature at Gargas are the numerous representations of mutilated 
hands, i.e. in many cases one or more finger-joints are missing. In various 
parts of the cave representations have been noted of one particular 
mutilated hand. This custom is also found among several primitive 
peoples, who cut off parts of their fingers as a sacrifice or to signify mour- 
ning. Representations of mutilated hands have even been found in caves 
in southern Australia. 

Besides these representations in colour at Gargas one may also find here 
another typical characteristic of the most ancient stage of artistic evolution: 
'macaroni', which are, of course, only to be found on the clay walls and 
roof of the cave. The surface of these arabesques cut in the clay has been 
faultlessly oxydized and covered with a delicate film, which makes it 
possible to distinguish them from engravings made at a later date. This 



52 



again illustrates distinctly the process whereby the first animal figures 
developed out of a tangle of lines; though still primitive, they are already 
inspired by a powerful sense of realism. A third group of figures comprises 
engravings where a burin was already used to cut the lines into the rock. 
From a stylistic point of view these representations of wild horses, ibexes, 
stags, oxen, bisons, mammoths and one marsh-bird are stylistically akin 
to the engravings in schist slabs excavated in Perigordian strata not far 
from the cave entrance. 

This extensive tunnel cave has become famous through the excavations hturiti 
carried out there by E. Passemard and the Comte de St. Perrier, who in 
the interior of the large north-west chamber found a great sequence of 
strata comprising all the cultures of the Upper Palaeolithic. In the 
Magdalenian strata in particular very fine portable art objects were 
discovered. In the centre of the chamber there stands an imposing stalag- 
mite pillar with bas-reliefs cut by Magdalenian artists into its soft tuff. 
According to the evidence of the deposits that partially cover these reliefs 
they derive from the early Magdalenian and Upper Solutrian. In a general 
survey which he has made Breuil draws attention to a reindeer facing left; 
below this is a horse, whose body was completely buried by the layer, and 
below this animal a bear facing left in the same manner as the horse. On 
the left of the frieze there is a mighty reindeer superposed by two smaller 
stags. 

This cave is situated to the west of the market town of Cabreret, over- Pech-Merle 
looking the Sagne, a small river. The exploration of this vast cave was 
undertaken between 1920 and 1922 by Abbe Lemozi and A. David, a local 
man. From the entrance, which was made at the time, the visitor soon passes 
a small low chamber, known as the 'ossuaire' (ossuary), in which the 
explorers came across the remains of bones of cave-bears. On the roof of 
this chamber very many faded silhouettes of human hands in red and 
series of red dots can be identified; there are also a number of 'macaroni' 
cut into the soft clay, and a very beautiful rendering of a large stag with 
impressive antlers. The 'great gallery' of Pech-Merle measures 153 yards in 
length and in some parts is 22 yards wide. Bizarre stalactite formations 
endow these chambers with a singular charm. It was here that Lemozi 
discovered the finest paintings and engravings. Noteworthy are three 
female figures of which copies were made by Breuil in 1924. They were 
drawn on the clay with the artist's finger; the women have pendulous 
breasts, their arms are summarily indicated, and they appear to be in a 
crouching attitude. Only one leg is shown, and in two figures the hair 
at the nape of the neck forms a queue. Not far away one can make out 



53 



an enormous animal between the arabesques cut into the clay, which 
Breuil has identified as either a bison or a musk-ox. Above it is the seated 
figure of a man with no head, presumably holding an arrow in his arm. 
Some 33 yards from these very early Aurignacian figures one comes to 
the most accomplished pictures in the cave. Here the south wall of the 
'great gallery' forms a hollow between two projections, known as the 
'chapel of the mammoths'. The drawings in black stand out with wonder- 
ful clarity against the delicate reddish background of the rock. There are 
about ten mammoths, four head of cattle, two or three bisons and one 
horse. Below these pictures there are groups of red dots undoubtedly 
originating from the Aurignacian. The animals belong to one of the final 
phases of the Perigordian or to an archaic phase of the Magdalenian. 



Frieze with black horses and black silhouettes of hands. Aurignacian-Perigordian. Pech-Merle. 

Width of frieze n ft. ii/4 in. 




54 



The twisted perspective occurs only in rudimentary fashion. The violent 
movement of a mammoth rearing in flight, seemingly seeking to escape 
from the deep abyss, has an affinity rather with the Perigordian style. The 
movements of the other animals are also portrayed in a vigorous realistic 
way, as at Niaux, Breuil is inclined to take the view that they were executed 
by an artist in an archaic style. 

Almost opposite this 'chapel of the mammoths' is the 'chamber of black 
hands', where the paintings are completely different from those just 
mentioned and must certainly date from an earlier period. The frieze is 
situated 104 yards from the modern artificial entrance and is 11 feet 
wide and 6 ft. 3 in. high. The visitor finds himself confronted with two 
large horses framed by broad black lines, their bodies filled in with 
black dots. Their manes are painted in uniform black and the heads are 
strikingly small. To depict the head of the animal on the right the artist 
utilized a natural relief in the rock. Both horses are shown in foal. 
Along the dorsal line of the animal on the right a pike has been drawn 
in red, its back studded with small red dots. Below the horses are remains 
of an older animal figure in red. The two horses are framed by six elab- 
orately executed silhouettes of hands in black. They represent left as well 
as right hands. According to Breuil this ensemble dates from the 
Aurignacian-Perigordian period. In another part of the 'great gallery' 
the silhouette of a hand in red, formed by 12 red dots, has been discovered. 
The small village of Sergeac is situated on the left bank of the Vezere, Sergeac 
not far from Montignac; close by, in a calc-spar, are the rock-shelters of 
Sergeac. The strata they contained were all from the Upper Palaeolithic 
and harboured paintings, engravings and sculptures. The rock-shelter 
of Blanchard, excavated between 1909 and 1911, yielded an inventory 
typical of the Aurignacian period, such as a boulder on which vulvae 
were incised; on another boulder are two bovidae outlined in black on 
a red ground. On the same side of the rock is the shelter of Castanet, 
in which several Aurignacian strata were unearthed. In the rock-shelter 
of Labattut several lumps of rock have fallen to the Perigordian level; 
they might therefore date from this period, but could equally well be 
older. In any case there is enough evidence at Labattut to permit the 
finds to be dated. On one of the boulders is a horse, represented in high 
relief; its four legs are suggested and it has a small head. Stylistically it 
brings to mind the pictures of horses at Lascaux. The finest figure is that 
of a stag of which only the head, neck and dorsal line have been preserved. 
The contours of the animal are drawn with superb skill in a flowing style, 
and the fine antlers are delineated in detail. There is undoubtedly a close 



55 



affinity between these paintings and those at Lascaux and the Cantabrian 
caves, as well as those of the Spanish Levant. 
Reverdit The rock-shelter of Reverdit, with its high reliefs, gives some useful 
guidance in determining the chronology of the Magdalenian period. The 
wall bears traces of a frieze which, however, has almost been destroyed 
by vegetation. One horse and three bisons may still be identified. Of 
greater importance are those parts of the frieze which are covered by 
deposit. Tools have been found here which may have been used to carve 
the reliefs. 
Laussel A few miles from Les Eyzies is the small chateau of Laussel, near which 
is the rock-shelter of that name. Laussel contained several Upper 
Palaeolithic layers and five large bas-reliefs on stone slabs were excavated 
from refuse of the Perigordian period. The famous Venus of Laussel is 
one of the finest female figures known to Ice Age art. This sculpture once 
adorned a block of calcareous rock in the rear of the rock-shelter, from 
which it has been hewn out and transferred to a museum. The breasts 
are fully developed, as is the case in all Aurignacian sculptures of this 
kind; the hips protrude, and at shoulder level she appears to be holding 
in her right hand an incised bison's horn; the left hand is stretched out 
across the stomach. The face, turned towards the horn, is round and devoid 
of detail, and the hair too is shown in outline, resting upon the shoulders. 
The figure was elaborately worked and stands out from the stone in full 
plastic relief. Traces of ochre indicate that it was once coloured red; 
this is borne out by the evidence of other female statuettes dating from 
the Perigordian era. 

Three other female figures found at Laussel are very similar to the 
Venus, but are somewhat smaller and have no horns in their hands. 
There is also a very beautiful male figure in profile, extraordinarily 
slender and shown without any genital organs. The man wears a girdle 
round his waist and appears to have once held an arrow or bow in his 
arms. 
Cap Blanc Only half a mile below the small chateau of Laussel is the rock-shelter of 
Cap Blanc, situated in a part of the country that offers abundant evidence 
of Ice Age art. In 1911 the discovery was made here of two Magdalenian 
layers superposed upon one another, both belonging mainly to the early 
Magdalenian. The rock-shelter, which is 49 feet long, harbours a magnif- 
icent frieze of horses as well as figures of bisons. This monumental frieze, 
which extends along a wall at the rear of the rock-shelter, constitutes a 
highlight of Ice Age relief modelling. The finest animals measure up to 
1 1 % in. in depth. The figures are executed in a wonderfully plastic manner; 



56 




Horse. Magdalenian. From the large frieze at Cap Blanc. 



unfortunately, however, some of them have deteriorated through weath- 
ering. The eyes are round and deeply incised; the bodies are fairly slender; 
the thigh-joints and parts of the chest provide distinct evidence of mod- 
elling; the manes are suggested by light hatching. Like the other bas-reliefs 
of the Dordogne and the Charente, that of Cap Blanc is dated by Breuil to 
the early Magdalenian. 

In 1904, in the course of an excavation in the small cave of La Greze, La Greze 
Dr. Ampoulange discovered in a hollow in the left-hand wall, hidden 
beneath deposit, the famous engraving of a bison. The animal measures 
2314 in. in length and is depicted entirely in profile, with the horns turned 
to the plane of the body. The outline is drawn sharply and distinctly; only 
one foreleg and one hind leg are visible, but no hooves. This work 
presumably dates from the Perigordian, although no typically Perigordian 
tool was found in the layer that covered it. 
This cave is situated not far from the small town of Le Bugue. It has been Barabao 



57 



known for a long time to the local population. The entrance is relatively 
large, but as one proceeds further the going becomes very difficult owing 
to masses of fallen rock and clay deposit. The paintings in the cave, which 
were discovered a short while ago, were examined in 1951 by Breuil and 
Windels, and later by Abbe Glory. With a certain amount of difficulty 
it is possible to identify some fairly large animal figures engraved in the 
soft clay walls with the artist's finger or a stick. There are 12 to 15 pic- 
tures, mainly of horses and oxen. Elsewhere a bison and a rhinoceros 
were deciphered. The figures vary in size between 3 ft. 3 in. and 6 ft. 
6 in.; they are drawn with rugged strokes but nevertheless display a 
vigorous naturalism. Breuil ascribes them to the Aurignacian-Perigordian 
period. 
Teyjat The cave of La Mairie is situated in the middle of the village of Teyjat, 
near Varaignes on the border between the northern Dordogne and 
Charente. In 1903 Peyrony went to Teyjat, where he discovered some 
magnificent engravings of animals on a yellowish stalactite column. Frag- 
ments of this pillar were unearthed during excavations made in late 
Magdalenian strata. At that time Breuil deciphered about 19 reindeer 
or remains of such animals, 10 stags, 3 head of cattle, 3 bisons and 2 bears. 
The cattle are arranged to form a scene: an ox follows a cow and behind 
the ox there trots a second ox. The outlines are drawn in the finest 
fluent Magdalenian style, individual details being carefully accentuated 
and the horns rendered in natural perspective. Some of the reindeer may 
be reckoned among the most magnificent engravings in Ice Age art; there 
is a delightful composition in which the female reindeer is rendered lying 
beside her calf. The bodies of the bisons are brilliantly portrayed, but 
the heads are drawn with curious indecision. It is gratifying to find that 
at Teyjat the pictures correspond in style to finds recovered from the layer. 
The figures at Teyjat are small, some no bigger than those found on 
portable art objects. 
Rouffignac The cave of Rouffignac, which stretches for more than 6 miles, c has for 
four centuries been an object of attraction to visitors, who have written 
their names in soot on its walls. In 1956 the owners of the cave, the Plassard 
family, invited two French prehistorians, L. R. Nougier and Romain 
Robert, to examine several figures which it was believed could be detected 
among the countless names. A large number of figures were discovered, 
either engraved or painted in black. On 17th July Abbe Breuil was also 
asked to verify their authenticity, and his verdict was positive. 
The entrance to the cave, which faces south-west, affords access to a long 
gallery extending in an east-west direction, the floor of which is covered 



58 



with a very slippery mass of mud (it has now been paved with stones). 
In an adjacent gallery, which runs from north to south and has many 
ramifications, the floor gradually becomes more solid. Here are the first 
of the extremely large number of pictures to be found in the main gallery 
and in the front part of several transverse galleries. They appear on walls 
and roofs. In certain parts the figures are amassed together. In this connec- 
tion mention may be made of two herds of mammoths facing one another, 
each with a male animal in the lead. Very close to a steep drop, by way 
of which one reaches a lower level of the cave, is a part of the roof 
with exceedingly fine paintings of mammoths, ibexes, bisons and rhino- 
ceroses. 

Unique in Ice Age art, in Breuil's estimation, are the three paintings of 
woolly-haired rhinoceroses discovered on the right-hand wall of the 
Breuil gallery. All these figures, which are outlined in black and are 

Head of rhinoceros, outlined in black. Early Magdalenian. Rouffignac. 




59 



sometimes accentuated by engraved strokes, are ascribed by Breuil to an 
early phase of Magdalenian. Whereas at Rouffignac representations of 
mammoths and rhinoceroses predominate, there are only few bisons and 
horses, and aurochs and cervidae are conspicuous by their complete 
absence — a fact presumably connected with totemistic differentiation 
between individual tribes. 

In one lateral gallery in the eastern part of the cave Nougier and Robert 
discovered a large section of the roof where the clayey surface is covered 
with countless 'macaroni'. In the front part of the chamber-like extension 
they are in an excellent condition, but their state of preservation deterior- 
ates as one proceeds further to the rear, until only traces of drawings done 
with the fingers are to be seen; the changes that have affected the surface 
of this part of the roof are attributed by Abbe Breuil to a warm current of 
air that was able to penetrate into the cooler interior of the cave through an 
old opening which is now sealed off by scree. Among the 'macaroni' very 
large representations of snakes stand out which recall similar figures drawn 
with the fingers at Baume-Latrone. The antiquity of the drawings indicates 
that this part of the cave was already frequented in the days of Aurignacian 
man. The number of works of art at Rouffignac is astonishingly large. Since 
Breuil discovered and examined the site 47 engraved and 31 painted 
mammoths have been identified (a figure that may well be subject to 
augmentation), and in addition 17 bisons, 11 ibexes, 10 rhinoceroses, 9 
horses and a few unidentifiable figures. Some of the pictures are as much 
as 6 ft. 6 in. in length. 
Pair-non-Pair This cave is situated above the right bank of the lower Dordogne, not 
far from Bourg-sur-Gironde. In 1883 F. Daleau observed engravings on 
the walls but paid little attention to them. Only after he had heard of 
similar discoveries at La Mouthe (1895) did he begin to take an interest 
in the works he had then observed. In 1896 he deciphered a horse that 
happened to be located in a position where the light was good. Excava- 
tions in the cave revealed a massive sequence of strata, more than 13 ft. 
6 in. thick, mainly consisting of Aurignacian and Perigordian layers, 
which completely covered all the engravings. One of the finest works to 
have come to light from the excavated part of the cave is a small horse in 
a frieze on the right-hand wall, lying with its head turned to the rear; 
it was called 'Agnus Dei' on account of the suggestive parallel with 
Christian iconography. It has delicately engraved contours, a small head, 
a large eye and a delicate mouth. The forelegs are clearly marked, whereas 
only one hind leg is shown. In another frieze, also on the right-hand wall, 
it is possible to identify a feline, facing to the right with the head shown 

60 



full-face; its hind quarters are superposed by an enormous mammoth, 
whose sharply sloping dorsal line is clearly visible; only one tusk is 
engraved In the small head. Nearby two bear's heads can be distinctly 
picked out. In addition to this right-hand wall there is a lateral gallery 
which harbours a wealth of engravings. The innumerable overlapping 
lines here are certainly confusing. The frieze is well illuminated by the 
daylight from the entrance. Superposed upon the engraving of a stag 
one can decipher a rhinoceros' head with a single horn and the turned-up 
snout characteristic of this animal. Next to it is another bear's head, and 
not far from this a horse with a sweeping neck-line. The engravings at 
Pair-non-Pair, which are exceedingly numerous, are ascribed to the 
Aurignacian-Perigordian culture on the evidence of the deposit that- 
covered them. 

In the course of excavations in the large rock-shelter of Roc de Sers, some Roc de Sers 
9 miles from Angouleme, Henri Martin discovered between 1927 and 
1929, one after the other, some very fine bas-reliefs hewn into several 
boulders. All these boulders were raised out of an Upper Solutrian layer, 
with the side that had been worked facing downwards; some of them had 
been smashed, and some boulders were even found on the slope in front 
of the rock-shelter, giving the impression that once in prehistoric times an 
iconoclastic storm must have raged here, as a result of which this place 
of worship was destroyed. Presumably the stone slabs were placed in a 
raised position on a semi-circular pedestal at the rear of the shelter. Among 
the animals represented it is easy to identify six horses, three or four bisons, 
partially incomplete, and several ibexes. The finest specimens are two 
male ibexes charging each other with lowered horns. On another stone 
slab a musk-ox is depicted chasing a small man carrying a stick across 
his shoulder. The horses are without exception shown with short legs and 
in foal. These reliefs, produced at the end of the Solutrian period, are 
masterpieces in no way inferior to the bas-reliefs at Cap Blanc or the 
equally fine frieze of horses at Chaire a Calvin (Charente) dating from 
the early Magdalenian. In this connection one may also mention a frieze 
widi bas-reliefs at Angles-sur-Anglin (Vienne dep.), discovered in 1949 



Fig. 12 — Ibexes butting one another. 
Bas-relief. Le Roc de Sers. Length 
of animal on right 1 ft. gl/2 •**■ 




61 




Bison, sculptured in clay. Magdalenian. Tuc d'Audoubert. Length of animal approx. 2 ft. 



by S. St.-Mathurin and D. Garrod; it yielded fine pictures of bisons, horses 
and ibexes, some of which were still on the wall whilst others had fallen 
into deposit dating from an advanced phase of the Magdalenian period. 
Noteworthy, too, here are the torsos of three life-size female figures. 
La Magdelaine This cave is situated on the right bank of the Tarn, not far from the town 
of Montauban. In 1952 M. Bessac discovered at La Magdelaine a fine 
rendering of a mare and one, less successfully executed, of a bison. Both 
animals may be ascribed to the Magdalenian period. The explorer also 
discovered two naked female figures, represented in a naturalistic style, 
facing each other on the right and left side of the chamber. According to 
Breuil these figures originate from the early Magdalenian. This early 
date is also suggested by fragments of other pictures surrounding the 
"Venus' figures. Breuil has drawn attention to the fact that these female 
figures are not present in deep caves where no daylight penetrated; all 
such examples, from the Perigordian figures of Laussel to these Mag- 



62 



dalenian figures at Angles-sur-Anglin, are to be found in rock-shelters 
which were once frequented. This leads Breuil to assume that Palaeolithic 
man did not represent "Venus' figures at places of worship. 

Not far from the railway-station of Gibaja, in the mountainous country CAVES IN SPAIN 
of Cantabria, lies the village of Ramales. Just over a mile above this Covalanas 
village are the caves of La Haza and Covalanas, situated in a broad rocky 
ravine. Both caves were discovered by Alcalde del Rio and Padre Sierra 
in 1903. From the entrance to Covalanas, which is 5I/9 yards wide, two 
galleries branch off, only one of which, however, can boast of paintings. 
These are to be found on the right-hand wall of the right-hand gallery, 
8a yards from the entrance. After two hinds in a very imperfect state of 
preservation one comes across a whole herd of these animals: one hind 
is depicted with its head turned to the rear, another has its head turned 
to the right, and a third follows immediately behind. All of them are out- 
lined in red and executed in a peculiar manner: the contours consist of 
spots, which in parts merge into one another. Presumably these were 
applied with a tampon. The three following hinds were painted in the 
same technique; one animal is caught very well in the act of flight. 
Another frieze comprises four hinds surrounding a horse with an elon- 
gated body. This manner of painting is a variant, somewhat blurred and 
flowing, of the linear stroke. Breuil assigns these paintings at Covalanas 
to a phase of the Cantabrian Perigordian. 

Some 9 miles from Santander is the small castle of Santian, near which Santian 
lies the cave of that name. It consists of one single gallery 224 yards 
long. The drawings are on the left-hand wall 142 yards from the en- 
trance. Here two rows of curious signs were painted in red. They appear 
to depict human arms and hands; others end in something rather like 
a trident, while other types again are club-shaped. Breuil assumes that 
these 15 signs originate from a very early phase of the Magdalenian pe- 
riod and that they may possibly bear some relation to the red signs on 
the roof of the picture gallery at Altamira; but it is also possible that 
they are connected with representations of hands. 

15 1/2 miles south of Santander, near Puente-Viesgo, there rises up a steeply- El Castillo 
sloping limestone massif, in which there are a number of caves. The 
most important one, El Castillo, was discovered by Alcalde del Rio in 
1903 and examined by Breuil three years later. The excavations at the 
entrance carried out by H. Obermaier and P. Wernert between 1909 
and 1914 proved fruitful, revealing a sequence of strata some 56 ft. thick, 
which yielded evidence of all the Upper Palaeolithic cultures. 



Proceeding through a large vestibule one reaches a spacious hall which 
continues to the left in several chambers; the floor shows numerous 
traces of frequentation by late Ice Age man. On the right-hand side 
of the gallery the visitor comes across engravings and paintings, as well 
as the well-known frieze of hands. The traces of scratches made by 
cave-bears before the advent of man are older than the representations 
of hands and the groups of dots to be found close beside them. During 
earlier ages daylight must still have penetrated into the large chamber, 
until this became impossible owing to the ever-increasing deposit, and 
in the Magdalenian period it was already shrouded in darkness. Breuil 
has put the number of well-preserved hand silhouettes at 44, 35 of which 
are of left and 9 of right hands. All the representations are surrounded 
by red pigment which has been blown on to the wall; none of them are of 
mutilated hands. 

Another group of representations comprises animal figures with linear 
outlines in red and yellow, which are to be found on the same parts of 
the wall as the renderings of hands. The former are often superimposed 
upon the latter and must therefore be more recent. It is, however, uncer- 
tain how much time elapsed between them. These animal figures are 
mainly of bisons, among which Breuil includes a horse and the head of 
a hind. The third group consists of paintings with outlines drawn with 
broader flowing strokes. Bisons painted in this style were also super- 
posed upon the silhouettes of hands. The fourth group comprises paint- 
ings in black, which are chiefly to be found at the rear of the chamber. 
The older paintings were executed with fine strokes, but later the strokes 
became broader, as is clearly shown by the hind's head and the horse. 
Many sketches of cattle and ibexes in black were drawn in cursory fash- 
ion but with a sure touch. The paintings are definitely early Magdalenian: 
to the late Magdalenian periods belong several bisons in black, where 
some attempt has already been made to model the bodies with paint. 
Two of the representations of bisons resemble the fine polychrome pictures 
at Altamira. The painter here made use of the red colouring of the faded 
hand silhouettes and added some black paint. With these efforts, however, 
polychrome painting at Castillo came to an end. 

The engravings in this cave are also of special artistic value: from the 
Aurignacian period we have the head of an ibex and several inferior 
pictures of horses. But magnificent Magdalenian engravings of stags and 
hinds, with the body surfaces filled in with hatching, are, however, a 
frequent feature at Castillo; they correspond completely to the portable 
art objects unearthed in the early Magdalenian layers. There are also 



64 



beautiful engravings of cattle and several bisons from the later phases 
of the Magdalenian period. 

This cave, which was discovered by H. Obermaier and P. Wernert in 191 1, La Pasiega 
is situated in the same massif as El Castillo, not far from Puente-Viesgo. 
The entrance in use nowadays leads into a small chamber whence gallery 
'A' leads deeper into the interior of the cave and is in turn connected with 
three other galleries, which in ancient times could be entered from out- 
side. Only the third gallery 'B' is of importance. It yielded several paintings 
of the Aurignacian-Perigordian era. The oldest find is probably a black 
painting of a hand of the positive variety. Engravings are rare in Pasiega 
and are not elaborately executed. Breuil describes red, yellow and black 
paintings of horses, hinds and bisons, followed by figures of hinds, horses 
and a stag outlined with red dots; he also lists several club-shaped signs 
and tectiforms. In Pasiega only two large bisons and a black stag seem 
to originate from the Magdalenian era. 

The entrance to the cave of Pindal, which is much exposed to rain and Pindal 
wind, is situated above the shore of the Gulf of Biscay, not far from the 
small village of Pimiango. From the entrance, situated on a rock terrace, 
access is gained to a large straight gallery, nearly 400 yards in length and 
varying between 11 and 22 yards in width. The first pictures are to be 
found 120 yards from the entrance. Almost all of them are scattered over 
the right-hand wall. One that is familiar is the elephant of Pindal: this 
is an enormous creature, rendered in a posture of repose, its contour fig. 13 
indicated by a fine red line; the forehead is high and bulging; the trunk 
hangs down, involute only at its extremity; the legs are long and the hooves 
mushroom-shaped. It differs from the Ice Age representations of mam- 
moths by having no long thick fur and by its shorter tusks. In the centre 
of the body there is a large red spot. Does this perhaps indicate the posi- 
tion of the heart? 

Nearer the entrance an engraving of a fish has been discovered, super- 
posed on a series of dots. Breuil has identified this as a tunny-fish and 
has ascribed it to an advanced phase of the Magdalenian period. Below it 
is a large engraved bison. To the right is a series of dots in red and black, 
arranged horizontally. Beneath a bison painted with a broad red line 
are six club-shaped signs. The style in which this animal is painted suggests 
late Magdalenian. 

The art station of Buxu is situated to the north of the village of Cardes Buxu 
in the Libas valley. The cave consists of a corridor 87 yards long. 66 yards 
from the entrance engravings have been found of horses, cervidae and 
ibexes. In Buxu there are also representations in black: 15 tectiforms of 



65 



m? 



F*&- *3 ~ Prehistoric elephant with heart paint- 
ed in. Pindal. After H. Breuil. Length i ft. }Y 4 in- 



a type found as well at Altamira. Breuil assigns the engravings and paint- 
ings at Buxu to the early Magdalenian. 
Pena de Candamo This cave was discovered in 1914 and studied by J. Cabre in 1915. 
From a spacious vestibule the visitor gains access to a point, still 
illuminated by daylight, where there are some painted red signs similar 
to those at Santian and Altamira. The entrance corridor now becomes 
narrower and descends fairly sharply to the interior of the cave. In 
a large chamber are a great number of engravings and paintings. 
Archaic animal figures recall the paintings at Covalanas and La Pasiega. 
At first they are executed with delicate strokes, later with wider red or 
dark lines; sometimes the contours of the bodies of the animals consist 
of single dots. The representations are of aurochs and horses. According 
to Breuil they originate from different Perigordian periods. Superposed 
on them are various engravings of horses, bisons and stags. The antlers of 
the latter are still rendered in twisted perspective. However, the technique 
of filling in and modelling the bodies with striation is somewhat reminis- 
cent of the engravings of the last stage of the Solutrian at Altamira and 
those of the early Magdalenian period at Castillo, and so we are presumably 
dealing with a retarded style — i.e., twisted perspective, an element of 
Perigordian art, is in this case found right up to the early Magdalenian. 
Los Casares Between the Cantabrian mountain range and the Mediterranean coast-line 
there are several caves, such as Penches and Atapuerca in the province of 
Burgos, which contain paintings of no importance. On the other hand, 
significant works have been found in Casares, a cave in the heart of Spain 
situated directly on the Aragon border. 

Casares, 214 miles to the north of the town of Riba del Saelices, was 
partially excavated by J. Cabre in 1934. He succeeded in deciphering a 
large number of very fine engravings in an advanced Perigordian style. 
Cabr£ recorded 15 horses, 10 aurochs, 9 stags, 4 ibexes, 2 lions, a rhinoc- 
eros, a glutton and a wolf. There are some older engravings with shallow 
incisions beneath these pictures which are engraved more deeply. Among 

66 



them are several striking anthropomorphous representations with gro- 
tesque faces partly resembling fish and partly toads. These creatures may 
perhaps have served a purpose in some water-rite: one of the figures seems 
to be diving into the water. Only a few of the pictures at Casares are paint- 
ed in black. 

The southernmost offshoots of Franco-Cantabrian art in Spain are to be 
found in the caves of La Pileta and Ardales in Andalusia, not far from 
Malaga. It was in 1911 that the Englishman W. Verner announced the 
discovery of the large cave of La Pileta in the Serrania de Ronda. This 
cave, like Baume-Latror.e, harbours very ancient rock art from the 
Aurignacian period. From multilinear arasbesques drawn with the fingers 



La Pileta 



Group of hinds, painted in red. Parts of the contours are stippled. Cantabrian Perigordian. 
Covalanas. 




67 



there developed the first attempts at animal paintings in yellow, red and 
black. The head of an ibex and of an ox have been deciphered. The 
representations in yellow are in several places superposed by other paint- 
ings and are the oldest finds in this cave. Somewhat more recent in origin 
are the figures with linear outlines not unlike those that frequently appear 
in Cantabrian caves. Breuil records in this group ibexes and hinds in 
yellow and cattle, horses and hinds in red. The numerous representations 
in this cave may originate from post-glacial periods. 

CAVES IN ITALY In 1950 the first engravings executed in the Franco-Cantabrian style to 
Levanzo be discovered in Italy were found in a cave on the small island of 
Levanzo, which belongs to the group of Aegadian Islands off the west 
coast of Sicily. These engravings are without exception well preserved 
and are to be found in the heart of the cave. They are covered with a 
dark patina. Here, too, Ice Age fauna are depicted as well as a series of 
curious anthropomorphous creatures, probably also in this case men in 
disguise performing a dance. 

The finest image of an animal, one of the best representations in the 
Fig. 14 whole of Ice Age art, is the rendering of a wild ass turning its head to 
the rear as though catching a glimpse of something; this is an immensely 
vigorous and animated engraving with lines cut deeply into the rock. 
The explorers also observed an ox following a cow, similar to that at 
Teyjat, a stag with undeveloped antlers, a small stag with very elongated 
forelegs and a poorly executed rendering of a small horse. As elsewhere 
in Franco-Cantabrian art, this southernmost offshoot manifests the same 
sort of superpositions, the same rather chaotic arrangement of paintings 
on the cave walls, and the same sureness of hand in applying a naturalistic 
style. 
RomanelH This cave, discovered as early as the beginning of this century by Baron 
G. A. Blanc, stands outside the framework of Franco-Cantabrian art, 
since in this case we have a semi-naturalistic style. Only the rendering of 
an ox bears a remote resemblance to Franco-Cantabrian animal pictures. 
The main find at Romanelli, which is situated not far from Castro-Marnio 
in Apulia, consists of stylized female figures, geometric motifs, bands of 
parallel lines, and scaliforms. Breuil deems it possible to see certain 
affinities with the old Perigordian figures in the Franco-Cantabrian area, 
an argument that seems to be corroborated by two engravings found on 
small stone slabs depicting a feline and a wild boar. 
Addaura In 1952 Signora Bovio Marconi reported the discovery of a small cave 
at the foot of Monte Pellegrino, to the west of Palermo. At Addaura, 



68 



Fig. i-f — Small wild ass. Levanzo. Height 9^/4 in. 



which is no more than about 20 feet long and 20 feet 
wide, sinter deposits broke loose from the walls as a 
result of the explosion of an American ammunition 
depot located in the shelter, thereby exposing the 
engravings beneath, which had hitherto lain hidden. 
Stylistically these pictures are akin to those at Levan- 
zo. Apart from two fine representations of cervidae 
and several horses rendered in Magdalenian style 
Addaura yielded a very animated group of human 
figures, depicted naked with some of them wearing masks. Two men 
appear to be engaged in a wrestling-bout, while two others lie sprawling 
on the ground. Both the discoverers and Breuil think it possible that these 
two men are committing suicide by strangulation. Their legs, which are 
bound together, fettered and bent far back, seem to be pulling tight a 
cord placed around the victim's neck. 

Whereas the animal figures at Addaura are undoubtedly related to Mag- 
dalenian art, the representation of the human figures approximates greatly 
to the human image as portrayed in the art of the Spanish Levant. 
Niscemie, another shelter to the south-west of Monte Pellegrino, harbours 
animal pictures in the same style as at Addaura. In Niscemie there are 
equine animals and an aurochs with horns drawn in perspective, which 
again points to the Magdalenian era. 

This selection of Ice Age cave art in France, Spain and Italy is designed 
to convey an impression of the advanced level of artistic achievement 
attained by Palaeolithic man in Europe. As already mentioned, it has not 
been possible to consider the minor arts in this connection. 
Towards the end of the Magdalenian period the ice slowly retreated to 
the Alps and the Arctic and with the changes that now occurred in the 
flora and fauna the post-glacial climate created new conditions for human 
existence. This, however, also eliminated the prerequisites for the nat- 
uralistic cave art that flourished during the Ice Age. The art stations sank 
into oblivion: roof falls, landslides and deposit blocked the entrances to 
the caves until, after the lapse of millennia, they were by chance once 
again opened to human eyes. 




69 



We now know of almost 120 stations of Ice Age art in the Franco- 
Cantabrian area. This figure, however, does not include all the caves in 
which early Palaeolithic paintings and engravings are to be found. The 
cave of Rouffignac in the Dordogne was only located a few years ago, 
and one may reckon with the discovery of more unknown art stations in 
the future. 



FRANCO-CANTABRIAN ART STATIONS 
WITH DATES OF DISCOVERY 



FRANCE Rhone Valley 

Ebbou (Vallon-Ardeche) 1946 

Le Colombier (Virac-Ardeche) 1947 

Oullins (Garn-Ardeche) 1951 

Chabot (Aigueze-Gard) 1878 

Le Figuier (Saint-Martin-Ardeche) \i 

Bayol (Collias-Gard) 1927 

Baurae-Latrone (Gard) 1940 

Salleles-Gabardes (Aude) 1947 



Lot 

Pech-Merle (Cabrerets) 1922 

Marcenac (Cabrerets) 1920 

Sainte-Eulalie (Cabrerets) 1920 

Cantal (Cabrerets) 1920 

Les Merveilles (Rocamadour) ic 

Murat (Rocamadour) 1914 

Cougnac (Gourdon) 1953 



Pyrenees District 

Aldene (Herault) 1926 

Niaux (Ardeche) igo6 

Bedeilhac (Ariege) 1907 

Ussat (Ariege) 1921 

Le Portel (Ariege) 1908 

Les Trois Freres (Ariege) 1914 

Tuc d'Audoubert (Ariege) 1912 

Mas d'Azil (Ariege) 1902 

Marsoulas (Haute-Garonne) 1897 

Montespan (Haute-Garonne) 1923 

Gargas (Hautes-Pyrenees) 1906 

Tibiran (Hautes-Pyrenees) 1951 

La Bastide (Hautes-Pyrenees) 1932 

Isturitz (Basses-Pyrenees) 1913 

Etcheberriko-Karbia (Basses-Pyrenees) 1950 



Dordogne 

Lascaux (Montignac) 1940 
Sergeac (St. Leon) 1909 
Belcayre (Thonac) 1934 
Laussel (Marquay) 1908 
Cap Blanc (Marquay) 1909 
Beyssac (Sireuil) 1915 
Nancy (Sireuil) 1915 
Commarque (Sireuil) 1915 
La Greze (Marquay) 1904 
Bernifal (Meyrals) 1903 
La Calevie (Meyrals) 1903 
La Mouthe (Les Eyzies) 1895 
Les Combarelles (Les Eyzies) 1901 
Font-de-Gaume (Les Eyzies) 1901 
Gorge d'Enfer (Les Eyzies) 1912 



70 



Oreille d'Enfer (Les Eyzies) 1932 
Jean-Blancs (Boumiquel) 1911 
Laugerie-Haute (Les Eyzies) 1864 
Laugerie-Basse (Les Eyzies) 1864 
Croze a Gcntran (Les Eyzies) 1913 
La Ferrassie (Les Eyzies) 1898 
Chateau-Ia-Tour (Sarlat) 
La Sudrie (Villac) 1937 
Barabao (Le Bugue) 1951 
Le Gabillou (Sourzac) 1940 
Teyjat (in Teyjat) 1903 
Fourneau d. Diable (Bourdeilles) 1919 
Rouffignac (near Rouffignac) 1956 

GrRONDE 

Pair-non-Pair (Marcamp) 1881 

Charente 

Roc-de-Sers (Sers) 1881 

Chaire a Calvin (Mouthiers) 1926 

Vlenne 

A ngles-sur- Anglin 

(near Angles-sur- Anglin) 1949 

La Marche (Lussac-les-Chateaux) 1937 

Yonne 

Arcy-sur-Cure 

(Grotte des Mammouths) 1946 

Tarn 

La Magdelaine (Penne) 1952 

Basque Provinces 
Berroberia (Pamplona) 1929 
Santimamifie (Vizcaya) 1916 
Venta de la Perra (Vizcaya) 1904 

Santander 

Sottariza (Gibaja) 1906 

La Haza (Ramales) 1903 



Covalanas (Ramales) 1903 

Salitre (Miera) 1903 

El Pendo (Camargo) 1907 

Santian (Puente Arce) 1905 

La Clotilde (Reocin) 1906 

Homos (San Felice de Buelna) 1903 

Altamira (Santillana) 1879 

Castillo (Puente Viesgo) 1903 

La Pasiega (Puente Viesgo) 1911 

Les Aguas (Novales) 1909 

La Meaza (Comillas) 1907 

Les Monedas (Puente Viesgo) 1952 

ASTURIAS (OviEDO) 

La Loja (Panes) 1908 

Pindal (Riba de Deva) 1908 

Mazaculos (Riba de Deva) 1908 

Quintanal (Balmori) 1908 

Buxu (Cangas de Onis) 1916 

El Cueto (Lledias) 1936 

San Antonio (Riba de Sella) 1912 

Bolado (Llanes) 1912 

Peiia de Candamo (San Roman) 1914 

Old Castile 

Atapuerca (Burgos) 1910 

Penches (Burgos) 1915 

Los Casares (Guadalajara) 1934 

Andalusia 

La Pileta (Malaga) 1911 
Ardales (Malaga) 1918 
La Cala (Malaga) 1908 
Palomas (Cadiz) 

Otranto 

Romanelli (Castro-Marino) 1904 

Siclly 

Levanzo (Levanzo I.) 1949 

Addaura (Palermo) 1953 



ITALY 



7' 



Rock art areas in the Spanish Levant 




1. Els Secans 

2. Val del Charco del Agua Amarga 

3. Gasulla gorge 

4. Valltorta gorge 

5. Morella la Vella 



6. Cuevas de la Arafia 

7. Alpera 

8. Cueva del Santo 

9. Cogul 



72 



THE ROCK ART OF 
THE SPANISH LEVANT 

BY 

HANS-GEORG BANDI 



The rock pictures of the Spanish Levant constitute the most vigorous 

works of art bequeathed by the prehistoric peoples of Europe. They 

are to be found in the hilly and mountainous hinterland of the coastal 

area that extends from the province of Lerida in the north to the 

province of Murcia in the south, or in other words from the Pyrenees 

to the Sierra Nevada. In contrast to the Franco-Cantabrian engravings 

and paintings they are not located in the interior of deep caves, hidden 

from human eyes: they are to be found beneath overhanging ledges of 

rock or shallow rock-shelters, in natural niches often situated at the foot 

of precipitous cliffs and separated from the bottom of the barren valleys 

by boulder-strewn slopes that can only be climbed with difficulty. Many of 

these niches are visible from afar, since owing to their ochre patina they 

stand out against the rest of the rock, which is generally grey. 

Many of the eastern Spanish paintings were long familiar to the local DISCOVERY 

population, and already before the turn of the century a note was AND STUDY 

published on one niche containing pictures situated near Albarracin 

(Teruel prov.). However, it was not until the beginning of the present 

century that they came to the attention of scholars. In 1 903 J. Cabre Aguilo, 

a photographer, came across several frescoes in the vicinity of Calapata, 

not far from Cretas (Teruel prov.): they comprise three stags and an 

aurochs painted in red. Cabre, however, did not realize the significance of 

his discovery until several years later, when he heard of the Ice Age 

cave paintings in the north of Spain. In 1907 news of this find reached 

H. Breuil, who at once took a keen interest in it. With this Spanish 

Levantine art came to the notice of international scholars, who without 

delay set about the task of studying it systematically. In this connection 

it was by no means unimportant that the controversy as to the authenticity 

and antiquity of the Franco-Cantabrian cave paintings had by this time 

already been concluded and that the significance of these Ice Age works 

had, after some curious false trails, at last been correctly assessed. 

From then onwards information about the rapidly mounting number of 

73 



discoveries made in this field was provided by several scholarly monographs 
and also by numerous articles and papers, particularly in the leading French 
specialist journal L' Anthropologic We can only go into this question quite 
briefly and have to rest content with singling out individual art stations. 
Cogul, situated to the south of Lerida, is an important site which was 
discovered at an early date; it houses the famous 'Dancing Women', a 
painting in red and black, about which the first scholarly work appeared 
as long ago as 1908. Even greater significance attaches to the find of rock 
paintings made in 1910 near Alpera, situated west of Almansa (Albacete 

Hunter running, and holding bow and arrows. The head, still visible in 1915, is hardly 
recognizable today. The figure was added at a later date to the neck of a large wild ox, but this 
animal has now almost entirely faded. Cueva del Charco del Agua Amarga, Teruel prov. Scale 

approx. 1 : 6. 




74 



prov.). Special mention should be made in this connection of a large 
painted frieze in the Cueva Vieja, which consists of a jumble of human 
and animal figures. In 1913 other important paintings were discovered at 
the Cueva del Val del Charco del Agua Amarga, near Alcaniz (Teruel 
prov.) and in 1914 at Cantos de la Visera on Monte Arabi near Yecla 
(Murcia prov.). 1914 was the year of the discovery of an important site 
containing many paintings near Minateda (Albacete prov.), on the rail- 
way-line between Madrid and Cartagena, and named after that town. 
These figures, of which there are several hundred, include a particularly 
large number of human beings on a frieze 60 feet long at the mouth of the 
Barranco de la Mortaja; according to H. Breuil they belong to 13 different 
periods of painting, so that this station possesses a special importance for 
the dating of eastern Spanish rock pictuies according to their style. 
In 1917 new paintings were found near Morella la Vella (or Vieja, 
Castellon prov.) and especially in the Valltorta gorge between Albocacer 
and Tirig (Castell6n prov.), some 25 miles from the coast. As in most 
other sites, the Valltorta gorge, a barren rocky valley, has not just one 
but several rock-shelters — 15 in all, which contain a large number of 
paintings. The most important are Cueva del Civil, Cueva de los Caballos 
(where the pictures, like many others in this region, appear to have long 
been familiar to the local population), Cueva Mas d'en Josep and Cueva 
Saltadora. The Valltorta gorge is one of the richest art stations of the 
Spanish Levant. The study of the various niches and their pictures is 
closely associated with the name of H. Obermaier. 

Among the discoveries during the following two years mention must 
be made of the pictures at Els Secans near Mazaleon (Teruel prov.) 
and at the Cuevas de la Arafia near Bicorp (Valencia prov.), where there 
is a famous scene of two honey-gatherers climbing up by means of a rope 
to the hiding-place of some wild bees (Fig. 29). Another delightful station 
is Tormon in the Olivonas valley near Albarracin (Teruel prov.), which 
was explored by H. Breuil and H. Obermaier in 1926; it contains a 
number of pictures of human beings, wild oxen, wild horses and stags, 
painted in red and black. 

Other sites with paintings of great significance were discovered in the 
early 1930s in the Barranco de Gasulla near Ares del Maestre (Castellon 
prov.), not far from the Valltorta gorge: the Cueva Remigia and the 
Cingle de la Mola Remigia, with several niches containing pictures. 
There are hundreds of human and animal figures, painted in red, black 
and brown, on a rock ledge comparatively high up the right-hand side of 
the barren valley. Here, too, H. Obermaier, in collaboration with H. 



Plate p. 78 



Plate p. 



75 




Fig. i S - Unfinished human figure. Paint- Breuil and J. B. Porcar, played a lead- 

in? in dark red. Valltorta soree, Castelldn ■ . • j- j i • 

i. ah t, sm. ■ c i , ._ ing part in recording and evaluating 

prov. After H. Obermaier. Scale approx. 1:3 ° r ° ° 

the pictures. In the course of his 
work in the Gasulla gorge, Porcar 
discovered not far away in the bar- 
ren little valley of 'Les Dogues' a 
singular scene of warriors in combat. 
No group of rock pictures as large as that found in the Gasulla gorge has 
been discovered since that time. But a number of smaller sites that have 
come to light during the past two decades show that the discovery of 
Spanish Levantine art is not yet at an end. Among the new sites we may 
mention the Abrigo del Arquero and the Cueva de Dona Clotilde near 
Albarracin (Teruel prov.), an area where, as has been stated, niches 
containing pictures had previously been discovered; also El Mortero 
and Cerro Felio near Alac6n (Teruel prov.), Cueva del Polvorin 
(Castelldn prov.) and several sites in the Barranco de Llort near Rojals 
(Tarragona prov.). We can take it for granted that further discoveries 
will be made in future. The number of sites on the Spanish Levant 
containing paintings so far mentioned in the literature on the subject 
totals about 50 — and in many places several niches have been listed under 
the same name. They are spread over the high-lying hinterland of the 
eastern Spanish coast, which suggests that this mainly rocky area was the 
home of one particular group of people. 
TECHNIQUE The majority of the works in the Spanish Levant are paintings; 
engravings are very rare. As a rule the paintings are monochrome, poly- 
chrome pictures appearing only in exceptional instances. The range of 
colours is limited: in most cases the prevailing colour is reddish (from light 
red to reddish brown), but black and sometimes white were also used. 
Natural pigments, such as manganese, haematite, limonite, ochre, red 
chalk and charcoal were employed. Tests carried out by K. Herberts show 
that the colours were probably applied in the form of a thin liquid. It 
was thus a glazing technique, whereby the pulverized pigments were 
apparently mixed not only with water but with a slightly sticky substance 
such as diluted blood, melted honey, albumen or vegetable juices, which 
acted as a binder. The paint must have been applied to the yellowish or 
greyish-blue rock in several stages with the aid of simple brush-like 
implements. An unfinished representation of a human figure at Cueva 
del Civil (in the Valltorta gorge) shows that the contours were apparently 
drawn first: we have here the outline of a leg, of which only a small part 
has been filled in with paint (Fig. 15). It is also clear from some pictures 

76 



Fig. 16 — Man with bow. Painting in red. Tormdn, 
Teruel prov. After H. Obermaier. Scale approx. 1:4-; 



that the inner surface was first paint- 
ed with a watery glaze, usually grey, 
and that the final colour was only ap- 
plied later, for the two coats do not 
always coincide completely. It is, of 
course, impossible to say whether this 
procedure was always quite so com- 
plicated, but in some cases it can be seen very distinctly. Finally, it may 
also be mentioned that sometimes — as, for example, in the case of a male 
figure from Tormon (Teruel prov.) — the inner surface is filled with a 
striped design (Fig. 16) instead of being painted all over in monochrome, 
and that in exceptional cases figures painted in two or more colours occur. 
On the other hand, there are several pictures in which the outlines are 
engraved. 

The preservation of the paintings in eastern Spain presents a particular 
problem. The fact that the pictures survived for several millennia — the 
question of their antiquity will be discussed below — is astonishing, 
considering that they were only to a very limited extent protected by 
overhanging rocks, and not everywhere at that. 

This can only be explained by the fact that the climate of eastern Spain is 
very dry. Another factor may have been that in the course of time a thin 
sinter deposit was formed by the percolation of water through the rock, 
which constituted a binding element similar to the calcium hydrate of 
the mortar in frescoes, and also formed a protective film. It is all the more 
regrettable that many of the pictures in this veritable 'archive' of art 
have suffered greatly during past decades, whereas generally speaking 
they were in an excellent state of preservation right up to the time of their 
discovery — apart from some minor damage caused by flocks of sheep or 
goats or by shepherds passing away their time. On comparing the present 
condition of the paintings with early photographs it can 
be seen that the reproductions are in general wonder- 
fully accurate, but, on the other hand, it is striking how 
many of them can scarcely be made out any longer. The 
reason for this undoubtedly is that the paintings are 
sprinkled, dabbed or rubbed with water by visitors 
to bring out the colours. The mere mechanical aspect 



Fig. ij — Archer. Alpera man. Painting in dark red. Cueva Saltadora, 
Valltorta gorge, Castellan prov. After H. Obermaier. Scale approx. 
1:2. 5 




PRESERVATION 




77 




Wild goat. To judge from the unnatural posture of the legs, the animal has probably been 
killed. Morella la Vella (or Vieja), Castellon prov. Scale approx. i : 3. 

of such treatment must inevitably cause damage, and in addition to 
this it enables dust to settle on the damp parts; but probably the most 
dangerous consequence of all is the chemical process that takes place 
when water is applied: the oxygen contained in the water seems to act 
on the limestone. Nowadays the state of preservation is unfortunately so 
bad in many places that the paintings are hardly visible unless the visitor 
sprinkles water on them himself, even though he realizes that this has 
a detrimental effect. Finally, it may be mentioned that the paintings 
have suffered greatly here and there from the actions of irresponsible 
persons and souvenir hunters. Unfortunately it is an almost impossible 
undertaking to try to prevent these abuses. Most of the niches in eastern 
Spain where paintings are to be found are situated in very isolated areas, 
and could only be protected by a railing, which would cost a great deal 
and probably be an eyesore as well, so that control and safety are not 
really feasible. This is not altered by the fact that only a small number 
of tourists go and see these paintings: experience teaches that the 'water 



78 



treatment' applied during the last Fig. 18 ~ Archer. Cestosomatic man. 

nr 1 j Painting in black. Cueva del Civil. 

fifty years or so has caused very f , 

' ' „ ■ Valltorta gorge, Castellon prov. Ajt- 

considerable deterioration; and anti- er fj. Obermaier. Scale approx. 1 -.4 

social elements feel safer in remoter 
places than at sites where they might 
be reprimanded by other visitors. On 
the other hand, the question of 
restoring certain paintings deserves 
to be examined. This applies, for 
instance, to the 'Dancing Women' at 

Cogul, where the figures can just be made out when the rock is sprinkled 
with water; we possess reliable information about the state of this work 
when it was discovered. Here and at many other sites expert restoration 
would surely produce results as rewarding as, for example, in the case of 
Roman frescoes. If nothing is done, in a few years' time we shall only be 
able to point out sections of rock where some beautiful dancing scene or 
other interesting painting could once be seen. 

The paintings of the Spanish Levant are mostly small, the individual 
figures being no larger than a man's hand. Human and animal figures are 
combined to form scenic compositions. This is the main difference between 
Spanish Levantine and Franco-Cantabrian mural art, since the latter, as 
is well known, consists mainly of naturalistic single animal figures, 
frequently drawn on a large scale. 

In the rock paintings of eastern Spain the animals are almost without 
exception depicted in a manner very true to life, but mostly very reduced 
in scale. If they measure as much as 30 in. in length, as is the case, for 
example, with some of the wild oxen at Torm6n (Teruel prov.), they 
may be considered quite large. With a good many of the animals their 
characteristic features are brought out distinctly, showing how keenly 
these hunter artists observed their quarry and how accurately they 
recognized their peculiarities. 

It is rather a different matter as regards the human figures: although they 
also display a naturalistic conception, they are frequently subject to very 
definite tendencies towards stylization. According to P. Wernert, at least 
four groups can be distinguished: the 'Alpera type', identified by its close 
adherence to nature and accurate proportions (Fig. 17); the 'cestosomatic 
type', with exaggeratedly long body, round head, broad, almost triangular 
chest, narrow hips and long, fairly thick legs (Fig. 18); the 'pachypodous 
type', with a comparatively short body, large head in profile, short slender 
torso and excessively thick legs (Fig. 19); and finally the 'nematomorphous 



79 




Fig. ip — Hunter. Pachypodous man. Painting in dark 
red. Cueva de los Caballos, Valltorta gorge, Castelldn 
prov. After H. Obermaier. Scale approx. 1:4 



type', with figures reduced almost to linear dimen- 
sions and stylized in such a way that their bodies 
are composed of no more than a few straight and 
curved lines (Fig. 20). This reduction or enlarge- 
ment of the body, or parts of the body, in the 
Expressionism manner of ombres chinoises, is frequently called Expressionism, and it is 
believed that artists who employed this method wished to suggest certain 
ideas of movement or force. But we must also reckon with the possibility 
that it was only at the beginning that the method of depicting human 
figures was consciously conceived and that subsequently the same methods 
were adhered to purely on traditional grounds. Nor can it be deter- 
mined conclusively whether these various types constitute different stages 
in the same tendency of stylistic development or whether they denote 
ethnographical or anthropological differences. In this connection it is 
noteworthy that no scenic compositions are known in which more than 
one of the types of human being mentioned appears. But in any case 
these figures, even where they intentionally or unintentionally depart 
to some extent from nature, can produce an extremely lively effect. Finally, 
it ought to be mentioned that in these human figures the face has a certain 
tendency to portraiture, and that importance is often attached to the 
rendering of head-dress, body ornaments, weapons, and — more rarely — 
other articles of equipment or clothing; we shall come back to this point 
later. 
SUBJECT-MATTER When the paintings are grouped scenically, which, as has already been 
stated, is frequently the case, they mainly depict the events of the chase. 
They portray the hunters and their quarry in a variety of situations. In 
one instance we see some men following an animal's tracks, which 
are clearly indicated; we may suppose that the hunters must have been 
very experienced, for the tracks cannot have been very distinct on this dry 
stony terrain. Sometimes the stalkers have already succeeded in bringing 
their quarry to bay and have just dealt the coup de grace. In pictures where 



80 



the hunter is shown running after game at full speed — as, for example, 
in the painting at Agua Amarga (Teruel prov.) — this appears to represent 
the chase. Particularly frequent are the pictures showing the battue: in 
many cases men and animals have been combined to form extremely 
animated groups. How magnificent are, for instance, the ibex or wild 
goat chase painted in red (Fig. 22) at the Cueva Remigia (Gasulla gorge), 
and the stag stalk at the Cueva de los Caballos (Valltorta gorge). In 
both cases a number of archers are lying in wait for the game, startled 
by invisible beaters, and overwhelming it with a hail of arrows. Whereas 
in scenes of the chase the representations of the running hunters are 
especially striking, here we are particularly impressed by the terrified 
animals, whose sole concern is to escape and who rush to their doom by 
coming within range of the arrows of the waiting hunters. The representa- 
tions of human beings here deserve attention, for the line-work of the 
lurking hunters, shooting their arrows from tautly-drawn bows, is in many 
cases very fine. The fact that hunting involved danger can be seen in 
pictures where the hunter is pursued by a wounded animal. One such 
scene (Fig. 23) is known to us from the Cueva Remigia (Gasulla gorge). 
Another interesting insight into the life of those who created eastern Battle-scenes 
Spanish art is afforded by representations connected with warlike episodes. 
Particularly delightful are, for example, the so-called 'advancing war- 
riors' from the Cingle de la Mola Remigia (Gasulla gorge): five men, some 
of them shown bearded, are marching one behind the other with a long 
stride, each holding several arrows in one hand and holding up a bow 
in the other (Fig. 24); this may represent a war-dance. Some other pictures, 
such as those from Morella la Vella (Fig. 25) and from the rock-shelter 
of Les Dogues, near Ares del Maestre, depict lively scenes of men engaged 
in a fierce battle, some of whom are wounded. Cueva 
Saltadora (Valltorta gorge) contains a picture, paint- 
ed in light red, of a warrior in flight: struck by 
several arrows, he collapses and thereby loses his 
head-dress (Fig. 26). Obermaier has called attention 
to certain representations from the Cueva Remigia 
to which a somewhat different significance must be 



Fig. 20 — Archer. Nematomorphous man. Painting in light red. 
Cueva de los Caballos, Valltorta gorge, Castelldn prov. After 
H. Obermaier. Scale approx. 1:2 




81 



Plate p. 96 



Plate p. 93 



Plate p. 96 



attached. The same motif, reiterated several times, depicts a group of 
men, drawn in a rather schematic manner, brandishing their bows as 
if expressing joy; on the ground before them lies an individual who has 
been struck by several arrows (Fig. 27). These seem to be regular scenes 
of execution, although it is, of course, difficult to determine whether 
those executed were prisoners of war or offenders belonging to the same 
tribe as their executioners, and whether they are being sacrificed or 
punished. In any case it is obvious that these pictures testify to the 
existence of definite juridical conceptions. 

The pictures so far mentioned by no means exhaust the subject-matter 
treated in the scenic rock pictures of eastern Spain. Mention has already 
been made of the honey-gatherers in the Cuevas de la Arana near Bicorp, 
who are warding off a swarm of wild bees (Fig. 29) and also of the so- 
called 'Dancing Women' of Cogul, where a group of women, some painted 
in red and some in black, are shown surrounding a small man; this picture 
may depict a ceremonial dance. At Minateda a mother can be seen walking 
and leading her child by the hand, and at Alpera there are two women 
who seem to be chatting to one another. There are several curious an- 
thropomorphous figures, half-human and half-animal, which may be 
identified as spirits of wild animals, or bush spirits, or possibly masked 
dancers. It is difficult to interpret some representations of insects, such as 
the spider painted in dark red, surrounded by a swarm of flies, to be 
found in the Cingle de la Mola Remigia (Gasulla gorge). 
The pictures also give us important information about the weapons, 
equipment, clothing, ornaments and hair-dress of the prehistoric inhab- 
itants of eastern Spain. By far the most important hunting weapon was 
undoubtedly the bow and arrow, depicted very frequently. Some of them 
appear to have been simple bows, as illustrated in the picture of a hunter 
at the Cueva Vieja, Alpera (Albacete prov.), or 'reflex' bows of various 
sizes (Fig. 28). Various types of arrow can also be distinguished according 
to their different heads and feathering. Spare arrows are sometimes carried 
in quivers, probably made of leather. It is likely that javelins were also 
used, but on the basis of these pictures it is difficult to distinguish them 
from arrows. A figure in the Cuevas de la Arana has been identified by 
K. Lindner as a lasso-thrower, but it may be a man brandishing his bow 
(Fig. 30). Other weapons and articles of hunting equipment have not been 
authenticated; there are also no indications whatsoever of fishing-tackle. 
But there are some pictures illustrating the use of containers and pouches, 
presumably made of leather or wickerwork (perhaps in some cases from 
clay). Noteworthy, too, is the use of ropes or leather straps for climbing. 



8a 







Archer taking aim. The bow, drawn taut with one arm, is only faintly recognizable in the left- 
hand part of the picture. Cueva del Civil, Valltorta gorge, Castelldn prov. Scale 1 :_?. 



S3 




Fig. 21 — Boar hunt. Paint- 
ing in dark red. Cueva del 
Vat del Charco del Agua 
Amarga, Teruel prov. After 
H. Obermaier. Scale approx. 
1 : 4 



SIGNIFICANCE 



The male figures are frequently depicted completely naked, but there 
are a few pictures (for example, that at Els Secans, Teruel prov.) in which 
a proper pair of breeches can be identified. In various places loin-cloths 
and waist-bands have been indicated, and one archer at Mas d'en Josep 
(Valltorta gorge) is depicted with a cape, the fringed ends of which hang 
down his back (Fig. 31). These articles of clothing must presumably have 
been made of beaten-out bark cloth, very fine wickerwork, leather or 
hide, since the art of weaving can scarcely have been known. We have 
a relatively good idea of the head-dress that seems to have been generally 
worn by men. In many cases feathers were either placed loosely in the 
hair or combined to form an ornament. Other figures are depicted wearing 
caps of various shapes, some resembling animals' ear-flaps; a fine example 
of this (Fig. 33) is to be found in the Cueva Vieja, Alpera (Albacete prov.). 
Quite often the men wear ornaments, mostly knee-rings, but also — less 
frequently — armlets; some of these must probably be interpreted as 
insignia of rank or honour. Finally, the paintings show that men wore 
their hair either short or hanging down to their shoulders; beards and 
moustaches were not infrequent. The women seem usually to have been 
dressed in bell-shaped skirts falling from their hips, whereas the upper 
part of the body was left naked; this is how they are represented perform- 
ing a dance at Roca dels Moros, Cogul (Lerida prov.) (Fig. 34). Some of 
the women adorn their arms with armlets or arm-bands; their hair hangs 
down loosely. 

From what has been said it follows that, seen as a whole, the paintings 
of the Spanish Levant (which, of course, do not consist exclusively of 
masterpieces but also include works of lesser merit) affords us a very in- 
teresting insight into this epoch of Iberian prehistory. But what was the 



84 



significance ,of these rock paintings? 
What motivated these hunter artists 
to produce paintings in some of the 
rock niches in their hunting-grounds? 
One explanation is that they were 
pictorial narratives, i.e. that certain 
incidents, such as important battles, 
successful hunting expeditions or 
memorable feasts were recorded for 
the benefit of contemporaries, or 
possibly of posterity. This idea should 
not be dismissed out of hand. And 
yet there are reasons for believing 
that matters are not so simple as all 
this. It is especially remarkable that 
these pictures are generally found con- 
centrated in one particular niche, al- 
though very similar rock formations, 
in some cases even more suitable for 
painting, were available close by. But 
while these show no sign whatever of 
having been used, the niches with 
paintings in many cases boast of a 
great number of works, which were, 
moreover, produced at different pe- 
riods and are therefore in part super- 
posed upon one another. What is the 
reason for the apparent preference 
given to certain places? How can one 
explain the mystery of their contin- 
uous use, in some cases for many gen- 
erations? M. Almagro argues that 
they were places of worship, sanctuar- 
ies which were constantly frequented 
because magic powers were ascribed 
to them or to the pictures they con- 
tained. He points out that at Cogul, 
for example, Iberian and Latin 
votive inscriptions suggest that the 
niches were for a long time famed 






Fig. 22 — Ibex hunt. Painting 
in light red. Cueva Remigia, 
Gasulla gorge, Castellon prov. 
After J. B. Porcar, H. Ober- 
maier and H. Breuil. Scale J .'5 



Fig. 25 — Hunter pursued by 
wounded wild ox. Painting in 
dark red. Cueva Remigia, Ga- 
sulla gorge. Castellon prov. 
After J. B. Porcar, H. Obermaier 
and H. Breuil. Scale 1:3 




?> 




»5 




From top to bottom: hind facing backwards; stag; wild goat. Second niche of pictures, Cuevas de 

la Arafia, Valencia prov. Scale i : 2 • 5 



86 



Fig. 24 — Warriors advancing (or dance scene?). 
Painting in greyish-black. Cingle de la Mola 
Remigia, Gasulla gorge, Castelldn prov. After 
J. B. Porcar' H. Obermaier a?id H. Breuil. 
Scale approx. 1 : 3 




for their supernatural powers. Curious- 
ly enough, it can actually be proved 
in some cases that the pictures were 
occasionally 'restored' in prehistoric 

times by the application of a new coat of paint. In this connection 
M. Almagro refers in particular to the bovidae at Tormon near Albar- 
racin and the 'Dancing Women' of Cogul. But there are, however, other 
paintings which have been subjected to minor alterations, such as the 
'stags' at the base of the large frieze in the Cueva Vieja near Alpera; 
here one can unmistakably trace older figures of wild oxen to which 
antlers have subsequently been added. 

We should also consider in this connection the existence of anthro- 
pomorphous figures, which must be identified as the spirits of wild animals, 
or bush spirits, rather than as masked men, and also the picture of spiders 
surrounded by small insects in the 
Cingle de la Mola Remigia, in the 
Gasulla gorge, and in the Cuevas de 
la Arafia near Bicorp. These are 
surely more than the product of some 
hunter passing the time while waiting 
for game to appear. 
All this suggests that we are not sim- 
ply dealing with pictorial narratives 



Plate p. 93 



c 



Fig. 25 — Archers fighting. Painting in dark 
red. Morella la Vella, Castelldn prov. After 
F. Benitez. Scale 1:1-5 



r-TS&i 



-</" 




r*c 



87 




DEVELOPMENT 



DATING 



Fig. 26 — Warrior collapsing after being struck 
by arrows. Painting in light red. Cueva Salta- 
dora, Valltorta gorge, Castellan prov. After 
H. Obermaier. Scale approx. 1:4-5 



about contemporary events. On the 
contrary, it seems that we have here, 
in some scenes at least, mythical in- 
cidents or conceptions derived from 
totemism. Possibly the hunting scenes are not renderings of one part- 
icular hunt, but the repetition of some primeval event. But, however this 
may be, the methods employed, the weapons, articles of clothing and 
ornaments represented in the pictures must correspond to a large extent 
with those in use at the time when the pictures were painted. The pictures 
thus do afford us an interesting insight into the life of the people who 
created Spanish Levantine art. 

In conclusion it should be stated that great caution is necessary in inter- 
preting the subject-matter of these paintings. It would be just as erroneous 
to regard everything simply as contemporary pictorial narrative as it 
would be to write all the pictures off as 'magic'. Only a thorough analysis 
of each scene can lead to plausible results. There is much of which we 
shall probably never be able to grasp the original meaning. 
The rock paintings of eastern Spain were not all produced at the same 
time; on the contrary, they must have appeared in some kind of sequence. 
This is shown not only by the fact that the paintings are frequently 
superposed, i.e. that the older pictures are partially or completely covered 
over by more recent ones, but also by the stylistic modifications that can 
be observed when one compares those of earlier and later date. On the 
basis of observations made by H. Breuil at Minateda, as well as his own 
experiences, H. Obermaier advanced the view that the eastern Spanish 
paintings show a trend of evolution, which led at first to ever greater 
perfection of style and technique but finally to decadence through exces- 
sive schematization. 

This brings us to the question of the antiquity of Spanish Levantine art. 
Briefly, the problem which has to be solved is whether the eastern 
Spanish paintings, like the Franco-Cantabrian ones, originate from the 
end of the Ice Age, or whether they are of more recent date, i.e. whether 
they belong to the Upper Palaeolithic or to a later cultural epoch. The 
main reasons why this question is more difficult to answer than in the 



Men with bows. Cingle de la Mola Remigia, Gasulla gorge, Castell6n prov. Scale approx. 1:2. 




case of Franco-Cantabrian art are: firstly, that the fauna depicted are 
mostly species that could live both in cold and warm climates; secondly, 
that we do not have any products of the artistic crafts, i.e. engraved or 
sculptured objects, corresponding stylistically to the mural paintings 
and found in strata of ascertainable date; and thirdly, that the eastern 
Spanish rock paintings do not occur in caves which can be proved to 
have remained intact from the end of the Ice Age until they were re- 
discovered in our own day. 






89 




Opinions are therefore very divided as to the antiquity and cultural at- 
tribution of the Spanish Levantine paintings. H. Breuil, H. Obermaier 
and a number of other scholars have pointed to the affinity with Franco- 
Cantabrian art and on these grounds are disposed to date them to the 
Ice Age and the Upper Palaeolithic. Several Spanish authors, on the 
other hand, have occasionally expressed the view that this is 'Neolithic 
art'. Nevertheless, this point of view has generally not been expressed in 
a categorical fashion, and these writers are often in two minds as to 
whether it is Mesolithic or Neolithic. An unequivocal opinion is voiced 
by M. Almagro, who in an excellent concise summary written in 1954 
states that these paintings date from a post-glacial period "in which they 
were definitely produced by Mesolithic hunters who for a long time retain- 
ed their backward way of life in the wild mountainous areas of eastern 
Spain." But he recognizes that there was a considerable overlap with 
Neolithic cultures. Despite this there is no need "to deny all points of 
contact and affinity which suggest that they were based upon Ice Age 
rock art. They could have been produced by Perigordian man, who 
belonged to a culture which was widely disseminated in Spain and formed 
the basis of the whole Upper Palaeolithic and Mesolithic, and which 

without doubt also lives on in the 
background of the present-day pop- 
ulations of the Iberian peninsula." 
The evidence of a considerable num- 
ber of excavations shows that towards 
j the end of the Palaeolithic offshoots 

of Franco-Cantabrian art reached 
eastern and southern Spain. This is 
borne out by the rock paintings in 
the cave at Pileta and two other sites 
in the province of Malaga, as well 



^*j5SSWht 




Fig. 27 — 'Execution'. Painting in red and 
black. Cueva Remigia, Gasulla gorge, Castel- 
lon prov. After J. B. Porcar, H. Obermaier 
and H. Breuil. Scale 1:2-7 



90 



Fig. 28 — Archer with reflex bow. Painting in black. Cueva Vieja, 
Alpera. Albacete prov. After J. Cabre. Scale approx. 1:2-$ 



as by products of the artistic crafts brought to light at 
Parpallo (Valencia prov.) and other caves. This affords a 
weighty argument against ascribing Spanish Levantine art 
to the Upper Palaeolithic. For it is more than unlikely that 
two different styles of Ice Age art should have co-existed 
side by side in eastern Spain, and not in any other region. 
Furthermore, it may be mentioned that the arguments adduced in support 
of the view that they date from the Ice Age are unconvincing. It is true that 
there do exist stylistic affinities between the animal pictures of the Spanish 
Levant and those of Franco-Cantabrian art which point to some possible 
connection. But this by no means implies contemporaneity; it can just 
as easily be a case of influences having been handed down from one period 
to another. The latter hypothesis is all the more probable since eastern 
Spanish paintings contain numerous other elements which are completely 
absent from Upper Palaeolithic art. 

The attempts made with the aid of certain animal figures to prove that 
the art of the Spanish Levant has its roots in the Ice Age are also highly 
arbitrary. Scholars believed that they could in some places identify 
figures of rhinoceroses, bears, lions, reindeer, bisons, wild asses and saiga 
antelopes. But these pictures are all of dubious antiquity, and M. Almagro 
has been able to show that in many cases the evidence rests upon sup- 
plementary drawings which differed in style from the originals to which 
they were added. There is also the possibility that isolated remnants of 
Ice Age fauna could continue to exist for a comparatively long time in 
the mountainous districts of eastern Spain; this is at least more than 
likely in the case of some of the animals mentioned, especially bears. 
And if the identification of the animals at Minateda as rhinoceroses 
is correct, they could well belong to the hairless African species, about 
which information may somehow have reached eastern Spain. Finally, 
one must also not exclude the possibility that animals may have been 
painted which were only known from legendary tradition. 
It is in any case certain that, wherever the identification of the species 
is quite clear, these animals lived in the Spanish Levant at the end 
of the Ice Age and that they could also exist there for a certain length 




9' 







DISTRIBUTION 



Fig. 29 — Two honey-gatherers climbing up a rope to 
the hiding-place of some wild bees and being attacked 
by them. Painting in red. Cuevas de la Arana, Valencia 
prov. After E. Hernandez-Pacheco. Scale 1:5 



of time in the post-glacial period, some of them even up to the present 
day. This applies chiefly to common stags, wild goats, rock deer, 
wild boars, wild oxen, and less frequently to wild horses, fallow deer, 
elks and chamois; birds and insects may be left out of account in this 
connection. 

Thus it cannot be proved that Spanish Levantine art dates from the 
Ice Age, and I personally am convinced that its origin goes back to the 
post-glacial era. We shall now attempt to define this period a little more 
closely. 

The subjects treated in these paintings make it quite plain that this is 
the art of a hunter people. To the argument that some figures or scenes 
(e.g. in the station of Villar del Huomo, Cuenca prov.) suggest that the 
domestication of animals was practised it may be said that all of them, 
like those presumed to date from the Ice Age, are open to question and 
cannot be identified for certain. And even if in some instances domestic 
animals are represented, this in my opinion by no means contradicts 
the view that we are dealing here with the art of hunters, since the occa- 
sional presence of such animals can be explained otherwise: they can be 
animals which the hunters had seen among neighbouring tribes, or had 
stolen or bartered from them. It is even possible that in this hunter culture, 
in response to external influences, the first steps were being taken in the 
keeping of domestic animals. As for the dog which seems to be depicted 
in several pictures, e.g. the stag chase in the Cueva Vieja, Alpera (Albacete 
prov.), it is not surprising that it should be present in a post-glacial 
hunter culture. We have authentic evidence that it appeared in the 
Mesolithic period in other districts (Maglemosian culture in northern 
Europe, Natufian culture in Asia Minor). 

Further evidence may be adduced from the distribution of Spanish 
Levantine paintings. We have seen that they occur almost exclusively 
in the mountainous hinterland of the coastal area and are not to be found 
in the strip of land along the coast itself. This suggests that this was an 
area to which the hunters in question were forced to retire by the advent 
of other peoples. This is also indicated by the fact that no discovery has 



92 



been made of any fishing scenes or renderings of boats, which is most 
remarkable in view of the proximity of the sea; we must at any rate 
reckon with the possibility that the coastal area was under the control of 
other tribes. Against this the argument may be advanced that the absence 
of such representations was due to the fact that the myths express- 
ed in these rock paintings for some reason or other bore no relation to 
the sea. But since the rock paintings certainly often manifest a mixture 
of mythical tradition and present experience, I attach no great signif- 
icance to this argument. 

Spider surrounded by flies. Cingle de la Mola Remigia, Gasulla gorge, Castell6n prov. Scale 
approx. 1:1-5. 




93 



Tn 



J 



j 



Fig. 30 — 'Lasso-thrower'. Paint- 
ing in red. Cuevas de la Arana, 
Valencia prov. After E. Hernan- 
dez-Pacheco. Scale approx. 1:2 





Thus we have now already obtained 
two clues to the age of our paint- 
ings: they appear to be post-glacial 
and to be the product of hunters. 
The question thus arises as to how 
long people with such an acquisitive 
mode of existence lived in the moun- 
tainous hinterland of the eastern 
Spanish coast. This cannot be easily ascertained: firstly, because we know 
relatively little about the Mesolithic cultures of this area; and secondly, 
because we have no reliable indications as to how quickly they came 
within the sphere of the Neolithic culture which established itself in the 
Iberian peninsula towards the end of the fourth millennium B.C. 
In the realm of eastern Spanish rock paintings various groups of Epipal- 
aeolithic finds have been discovered. Some of them were excavated strati- 
graphically in the area of, or close to, niches containing pictures — for 
example, at Cueva de Dona Clotilde, Cocinilla del Obispo and Prado 
del Navazo near Albarracin (Teruel prov.), Cueva de la Rabosa or de 
los Melones in the Valltorta gorge, and Roca des Moros near Cogul (Lerida 
prov.). Others were found at open stations. They manifest the tendency, 
characteristic of the Mesolithic period, to produce, in addition to rather 
badly-worked flint implements of normal size, small stone objects, often 
geometrically-shaped, which are mounted in frames of wood or bone. 
It is more than probable that the people who produced these stone im- 
plements may also be regarded as the 
creators of the Spanish Levantine 
pain tings. They may simply be descend- 
ants of the Upper Palaeolithic pop- 
ulation of the same area. But we 
must also reckon with new influences 
introducing fresh elements, the most 
likely area from which they could 
have originated being that of the Cap- 
sian culture in North Africa. These 
innovations could in theory include 
not only the differently-shaped mi- 
crolithic flint implements, but per- 
haps also the ideas expressed in the 
paintings. When these influences 
began to exert themselves upon the 



Fig. 31 — Archer with shoulder- 
cape and loin-cloth or girdle. 
Painting in dark red. Mas d'en 
Josep, Valltorta gorge, Castelldn 
prov. After H. Obermaier. Scale 
approx. 1 : 2- ; 



94 



indigenous Epipalaeolithic or post-glacial tribes cannot as yet be deter- 
mined. It may have taken place some time between 8000 and 5000 B.C. 
But one would have to make a thorough study of all the Epipalaeolithic 
material, supplemented by further finds, in order to establish whether 
the external influences appeared at once or gradually. The compilation 
of accurate maps could perhaps also show whether the Epipalaeolithic 
and Mesolithic hunters lived at first in the proximity of the coast and 
were then gradually pressed back into the mountainous hinterland, and 
whether an explanation based on cultural differences can be given for 
the absence of analogous pictures in the south of the Iberian peninsula. 
Without these essential prerequisites it is difficult to say anything about 
the date when paintings appeared for the first time in Spanish Levantine 
style. If this genre has its roots partly in the art of the last phase of the 
Ice Age (we shall come back to this point later), it must be assumed that 
it is very old indeed. On the other hand, its location in the mountainous 
hinterland seems to suggest that it developed at a relatively late date. 
The almost complete absence of clothing in the pictures of human beings, 
to which we have already referred, also seems to indicate a markedly warm 
period (climatic optimum 4th-grd millennium B.C.?). In this connection 
it may be added that according to some scholars the bell-shaped skirts 
which can easily be identified in pictures of women presuppose not only 
the use of home-made (or, more probably, bartered?) woven cloths, but 
may possibly also point to influences radiating from the Minoan civiliza- 
tion in Crete. This, however, would suggest a surprisingly late date. We 
are just as much in the dark as regards the date when eastern Spanish 
art came to an end. For we do not know whether rock art continued until 
the end of the old hunter culture; nor do we have any sure indication as 
to the date when the Mesolithic population became 
assimilated or was wiped out by the cave or Almerian 
cultures, or by a Neolithic facies derived from them, 
as they gradually spread towards the interior. Pre- 
sumably this took place in the main some time during 
the 3rd millennium B.C., with remnants surviving 
for a little longer. 



Fig 32 — Man with knee-breeches. Painting in dark red. Els 
Secans, Teruel prov. After H. Obermaier. Scale approx. 1:) 




95 




Below, on the right: an anthropomorphous figure (possibly the spirit of a wild animal or a man 
with an animal mask); above, on the right: wild goat (?); below, in the centre, a receptacle (a small 
basket or bag) over six horizontal arrows; above, on the left, man with bow. Cingle de la Mola 

Remigia, Gasulla gorge, Castell6n prov. Scale i : i -8. 

ORIGIN A final point to be considered in connection with Spanish Levantine art 
is the question of its origin. We are faced in particular with the following 
problem: do any connections exist with Franco-Cantabrian Ice Age art, 
or could an individual rock painting style have developed in the 
Epipalaeolithic and Mesolithic cultures of eastern Spain independently 
from the Upper Palaeolithic tradition? And, irrespective of the question 
whether they were based on ancient traditions or not, did external 
influences play a part in the development of Spanish Levantine art? 
It has already frequently been pointed out that the paintings of the 
Spanish Levant bear some resemblance to the Franco-Cantabrian pictures. 
This is in fact the case with some of the animal pictures. Consider, for 
instance, the stag at Labattut (Dordogne dep.), which we have had 
occasion to mention, and which may possibly bear comparison with one 
of the cervidae at the Cueva Remigia (Castell6n prov.); or the wild 



96 



oxen at Lascaux (Dordogne dep.), F'g- 33 — Ma " u ' ith b °v a "d 

which distinctly resemble those at «*; Pai " tin S in ™ d - Cueva 

Vieja. Alpera, Albacete prop. 

Roca dels Moros near Cogul (Lerida After H.Obermaier. Scale i : 4 ■ 5 

prov.). But these similarities are 
rather of a general nature and only 
concern animal figures. In spite of 
this one would not be wrong in 
assuming that Upper Palaeolithic 

traditions played their part if one knew for certain that Spanish Levan- 
tine art had already appeared in an early phase of the eastern Spanish 
Mesolithic; all the more so since there is authentic evidence of Franco- 
Cantabrian minor arts in the eastern Spanish area, most notably the 
painted and engraved stone slabs from the cave of Parpallo (Valencia 
prov.). If this were not so, we should have to assume that traditions which 
had fallen into oblivion for a long period of time suddenly burst into 
life again later, which is highly improbable. 

It is, however, certain that, animal figures apart, eastern Spanish rock 
painting follows an individual style markedly at variance with that of 
earlier Ice Age art. In theory, the new elements — i.e. the entire art 
of the Spanish Levant — may possibly have developed independently on 
the spot. But before we discuss this point we must still deal with the 
question whether any links can be established with other areas. 
Apart from several rather peculiar human figures at the cave of Addaura 
(Monte Pellegrino, near Palermo, Sicily) there is nothing in Stone Age 
Europe that could be related to Spanish Levantine art. It is therefore 
necessary to look towards Africa. We have already mentioned (in connec- 
tion with the question of the origin of the microlithic element in the 
inventory of post-glacial stone implements in eastern Spain) the prob- 
ability that African impulses could have played at least some part in the 
development of the eastern Spanish style. There are, moreover, doubtless 
some resemblances between the paint- 
ings of the Spanish Levant and 
certain rock pictures on the Dark 
Continent. According to K. J. Narr 
a more recent pictorial style can be 
traced from South Africa through 
Southern Rhodesia and East Africa 

to eastern Spain. The same author Fig. 34 — Two figures from the 

therefore speaks of a 'Eurafrican 'Dancing Women'. Painting in 

. , , , , black. Cogul, Lerida prov. After 

hunter culture, the last remnants ]. Cabri. Scale appr ox. 1:8 




> 




97 



* 



98 



of which can be found today with the Bushmen and some other non- 
Negro peoples in Africa. There are, indeed, as has been observed long 
ago, startling resemblances between the eastern Spanish paintings and 
those of South Africa. The great distance which separates these two areas 
can to some extent be reduced by finds in the same style in Rhodesia, 
East Africa, Egypt and the central Sahara. But there is as yet no convincing 
evidence of a link across North Africa; although this region, too, can 
boast of extensive centres of rock paintings, they differ from those in 
other regions. Nevertheless, there is a certain probability that connec- 
tions do exist, despite the distance involved, between the Iberian penin- 
sula and the southernmost parts of Africa. We have not as yet any detailed 
information about these connections, which can only be identified 
indistinctly. Only a series of stylistic comparisons will make possible some 
progress in solving this question. Serious difficulties also exist in that the 
dating of the relevant pictures in rock strata in Africa is just as uncertain 
— if not more so — as the dating of the Spanish Levantine paintings. 
Consequently it is not easy to answer the question whether the style 
of eastern Spanish rock pictures developed on the spot or whether it can 
be traced back to African impulses. For it is just as feasible that the 
influences radiated from the Iberian peninsula to Africa as from south 
to north. The circumstance that the Mesolithic groups of eastern Spain 
were probably influenced by the Capsian culture of North Africa is of 
no particular significance; in any case, there are as yet no indications 
that a new style of painting was introduced to the Spanish Levant from 
the Capsian area. Nevertheless the possibility remains that the basic ideas 
of eastern Spanish rock painting have their roots in Capsian culture. It is, 
however, strange that similar paintings are non-existent in the south of 
the Iberian peninsula. Must we therefore reckon with influences in- 
troduced into the Spanish Levant not by way of Gibraltar but directly 
from North Africa? 

Thus we see that connections between eastern Spain and the African 
centres of rock art are quite probable. But it is still an open question 
whether the impulses came from the Iberian peninsula or from Africa. 
The origins of Spanish Levantine art are therefore still shrouded in 
mystery. Further research is required before this problem can be solved. 



F'S- 35 ~ Stag hunt. The animal to the 
left of the stag may 'be a dog. Painting 
in black. Cueva Vieja, Alpera, Albacete 
prov. After J. Cabre. Scale approx. 1:8 




THE ROCK ART OF THE MAGHREB 
AND SAHARA 

BY 

HENRI LHOTE 



By the Maghreb we understand Morocco, Algeria and Tunisia, the region 
once called by the Arabs 'the island of the Maghreb'. The backbone of 
the Maghreb is the High Atlas mountain range. The most important 
specimens of rock art are to be found south of the Atlas mountains, and 
those situated in the north are only of secondary importance. 
By the Sahara we understand the desert area extending south from the 
Saharan Atlas to the glass steppes of the Sudan. Our region of study also 
embraces the Fezzan, Tibesti, Borku and Ennedi regions, although the 
last-named district forms part of the Sudanese steppe. We shall not 
consider here the deserts of Egypt and the former Anglo-Egyptian Sudan, 
although these regions also contain rock engravings. 
The first discoveries of prehistoric works of art were made in the Maghreb 
and the Sahara at a time when the origin of mankind was still in fierce 
dispute and no one dreamed of the existence of prehistoric art. In 1847 
officers of General Cavaignac's expedition against the Ksour tribes .in 
the mountains of southern Oran chanced to come across large rock 
engravings at Thyout, Moghar and Tahtani. The figures depicted 
elephants, lions, antelopes, bovidae, ostriches, gazelles, he-goats and 
human beings armed with bows, several of whom were linked to one 
another by their genital organs. The discoverers had no doubt that these 
were ancient works, dating from an era before the Arab invasion but 
after the time of Carthage. They concluded that they were the work of 
idolaters believing in fetishes, who had been brought to the oases of 
southern Oran in caravan expeditions from the south of Africa; they 
assumed that the artists must have been Tuareg. For the first discovery 
of rock drawings in the Sahara we are indebted to the German scholar 
Heinrich Barth; in 1850, during the course of the famous journey that 
was to take him from Tripoli to Timbuktu by way of the Fezzan and 
the Chad, he discovered near Tel Issaghan in the Fezzan several rocks 
bearing engravings of human beings with animal heads and a herd of 
cattle. When, following the track along the Air massif, he reached the 



DISTRIBUTION 



DISCOVERY 



99 




CENTRES OF ROCK ART 
IN NORTH AFRICA 



springs of Isolane in the southern part of the Anahef he found further 
animal engravings and written characters, the products of the Tuareg. 1 
Barth recognized these engravings as of high quality, the handiwork of 
skilled draughtsmen. He believed that they showed affinities with 
Egyptian art and that they must have originated from the Carthaginian 
epoch. Already at that time he also came to the interesting conclusion 
that the animals depicted on the rocks bore witness to the fact that the 
climatic conditions were once completely different from those of the 
present era, and that the population of those days must have been familiar 
with cattle-raising. 

In i860 a French scholar, Henri Duveyrier, found inscriptions in the 
Tassili and reported that, according to information supplied by the 
natives, there were other engravings, representing carts drawn by oxen, 
near Anai in the Fezzan. 

In 1869 Dr. Nachtigal explored the Tibesti, where he came across engrav- 
ings of cattle: he identified these as animals which the natives had seen 
in the south, and did not ascribe them to an early period. In 1887 Erwin 
von Bary mentioned rock engravings, representing human beings, horses 
and camels, near Dokou in the Air massif. 

Finally, the explorer F. Foureau, during the course of various journeys 
undertaken between 1893 and 1899 in the central Sahara, as well as in 
the Air massif, discovered a large number of engravings — both in the 
Tassili and in the Air massif. 

Already at that time it was recognized that the rock engravings were 
distributed over a wide area and could be found in all the larger massifs. 
This fact was to be corroborated by numerous scholars, travellers and 
officers who visited the region once the occupation of the Sahara by the 
French and the Fezzan by the Italians made it safe to do so. In southern 
Oran, in addition to the rock drawing sites at Thyout, Moghar and 
Tahtani, others were found at Bou Alem, El Richa, Mouchgueug, 
Chebkha Dirhem and elsewhere — in other words, the whole region 
from Beni-Ounif to Aflou can boast of very fine large engravings. 
The first important discoveries in the Sahara were made in the Hoggar, 
whereas the great finds of rock drawings in the Fezzan were only later 
discovered by Italian explorers and analysed by Frobenius. In this region 
the discovery was made for the first time of representations of war chariots 
drawn by horses; these were new elements which had no connection with 
the finds of engravings made earlier and must be dated to a more recent 
period. 
Whilst performing his military service in Ahnet, the western part of the 



Change of 
climatic conditions 



Th. Monod 



Hoggar, Professor Th. Monod discovered, over a period of 18 months, 
a large number of engravings which he subjected to careful study. The 
publication of the results of his investigations furnished valuable material 
of fundamental importance for all later research. 

Several years later a French lieutenant named Brenans discovered in the 
Wadi Djerat, a valley in the Tassili-n-Ajjer massif, a vast complex of 
engravings which apparently showed analogies with the fine pictures of 
the Fezzan and southern Oran. Among them were war chariots superior 
in style to those of the Fezzan. There were overhanging rocks covered 
with paintings from the same period, depicting chariots, human beings 
and animals. 

After Nachtigal a geologist named Dalloni recorded in the Tibesti a 
large number of other engravings, while British archaeologists accom- 
panying Prince Kemal el Dine found engravings and paintings, mainly 
of cattle, at the Djebel Ouenat. Later Professor Th. Monod and some 
French officers were to discover a further vast quantity of engravings and 
paintings there. 

On the Ennedi plateau, too, paintings were found, and not long ago some 
engravings as well. 

A most useful detailed inventory of the rock engravings of this massif has 
just been completed by a colleague of mine, M. Bailloud. 
The exploration of Upper Egypt and the Sudan likewise yielded nu- 



+ 




o 
O 



o 



Fig. 36 — Horse with human figure in double tri- 
angle style and Libyco-Berber characters. Tit, Hoggar 



102 



merous specimens of engravings and hieroglyphs dating from different 
periods, of 'which some definitely originate from pre-Pharaonic times. 
The first discoveries of rock engravings in the west were reported from 
southern Morocco in 1875 by Rabbi Mardochee ben Serour; the figures 
were of elephants, rhinoceroses, horses and human beings. In 1884 
Dr. Oskar Lentz made similar discoveries in the region of Foum el 
Hassane. Since that time exploration has been carried on by French 
archaeologists and we now know of a large number of sites containing 
rock pictures. In the Atlas mountains of Marrakesh engravings of chariots 
were found on Talaat-n-Iisk at a height of over 7000 ft. 
In Rio de Oro Spanish explorers have discovered engravings akin to those 
in southern Morocco. 

At Zemmour horizontal shelves of rock have been found covered with 
small engravings of no great antiquity; at this spot 107 schematically 
drawn chariots have been recorded. 

The exploration of Mauritania has been carried out by Professor 
Th. Monod, who has registered 1846 engravings and several paintings. 
Paintings were also discovered in the Adrar region of Mauritania by 
Mile, du Puigaudeau and Mile. Marion Senones. The Air massif, situated 
south-east of the Hoggar, was thoroughly explored by the Englishman Fir.. 36 
Fr. Rodd, who discovered some particularly fine engravings of horses 
and camels as well as Libyco-Berber hieroglyphs. Other engravings from 
this area were brought to notice by M. Nicolas. 

The massif of the Adrar des Iforas, situated south-west of the Hoggar, 
yielded even more finds than the Air. Here a number of sites were 
identified by some officers, as well as by M. Zohrer of the Ethnographical 
Museum at Neuchatel, Switzerland. 2 

To sum up briefly, all the mountainous regions of the Sahara contain 
rock engravings or paintings — often both together — and today, although 
an inventory of them has not yet been completed, we know of more than 
30,000 engravings; in the Tassili alone there are more than 15,000. No 
other part of the globe can boast of such an important and varied wealth 
of rock art as this part of Africa, which paradoxically enough is at the 
same time the largest desert on earth. 

Already at the time of the first discoveries it became apparent that not dating 
all the engravings originated from the same period. It did not take long 
to realize that there were marked differences in style, technique and 
patina, and that in many groups of pictures we find animals depicted 
which are nowadays extinct in the Sahara, such as elephants, rhino- 
ceroses, large buffaloes (bubalus antiquus) and cattle, while in others we 



103 




Group of rhinoceroses and ostriches. The inner surface of the figures is polished. Semi-naturalistic 
style. Bubalus period. Wadi Djerat, Tassili. Length 4 ft. 10I/4 in. 



104 



find specimens of present-day fauna such as horses, camels and moufflons. 
Striking variations are also evident in the portrayal of human beings, 
who are, moreover, armed with different kinds of weapons: axes, boom- 
erangs, bows, javelins, swords and fire-arms. 

The techniques of line engraving are particularly varied, and a study of 
these can in itself enable us to draw valuable conclusions. 
In the case of the oldest engravings, i.e. those with a very dark patina, 
the contour is usually formed by a deeply incised line, a groove which 
in cross-section has the shape of a levelled-off V or a U. This groove can 
be almost half an inch wide and is polished quite smooth; it is not always 
regular, and the width varies as well. The degree of irregularity of the 
contour denotes the quality of the engraving and also shows that the 
tool used for polishing must have been rather uneven — or, in other 
words, that this was no mechanical operation and required great skill. 
In groups of engravings treating the same subject there are figures of 
which the contours are not polished but pecked out. Both techniques 
were employed simultaneously, for unfinished figures show that engrav- 
ings with polished contours were first pecked out. This technique of 
preparatory roughening is still used today by stone-masons, who before 
setting about polishing marble or granite chip it with a hammer to obtain 
small indentations and give their polishing tools or machines a better grip. 
On the prehistoric engravings the pecked lines consist of rows of small 
conical holes, generally not connected with one another but separated 
by fairly small spaces. As a rule it can be said that where these holes are 
closer together this makes for greater regularity and beauty of form. In 
some engravings the whole area within the outline, or part of it, is also 
pecked out and polished. 

The techniques of polishing and pecking were employed in several epochs 
by artists following different styles, but it is easy to establish that the 
older works are superior and that a gradual decline in quality occurred; 
the most recent engravings show signs of rough chipping and are no 
more than mediocre as works of art. 

According to some authors the reason for the greater perfection of the 
oldest engravings is that the artists of that epoch were more skilled in 
working stone; technical quality would thus act as a criterion of a work's 
antiquity. 

No tool has as yet been found which could provide an explanation of the 
technique employed at the time in pecking, and more particularly in 
polishing. The finest engravings are to be found on sandstone rock, which 
is undoubtedly more easily worked than harder types of stone, although 



Technique 



105 



quartz and granite are also found bearing engravings. Striking with a 
hard pointed stone can produce small indentations in the rock, but not 
so deep, and — more especially — not so regular, as is the case with the old 
engravings. One pointed stone must therefore have been used as a chisel 
and another one as a hammer. This hypothesis is borne out by the fact that 
striking tools of a Neolithic type are frequently to be found close to 
some engravings, whereas this spot cannot have been used as a depot. 
This is, however, not always the case; not a single Stone Age tool, for 
instance, was found near the large assemblages of old engravings in the 
Wadi Djerat in the Tassili. 

The polishing of the groove is an even greater puzzle, for it is difficult 
to imagine what kind of tool can have served this purpose. Some grooves 
with a V-shaped cross-section suggest that the tool used may have been 
the cutting edge of a Neolithic axe, with the aid of damp sand. But the 
grooves in which the cross-section is levelled off in the shape of a U, as 
is the case with the finest engravings, cannot have been produced with 
any of the stone implements known to us. Had stone axes been used, to 
make either V- or U-shaped grooves, specimens of them with appropriate 
traces of wear would inevitably have been found close to the sites of the 
engravings. But this is not so. Possibly the artist used a pointed stick of 
very hard wood and damp sand to polish the grooves after pecking them. 
But in these questions we have to resort to hypotheses, since no tools have 
been found which could help to determine the date. 
Patina Another criterion for establishing the antiquity of the Saharan engravings 
is their patina. It has been ascertained that with some engravings the 
colour of the contours is identical with that of the rock worked, whereas 
with others they are lighter, and indeed are found in a whole range of 
shades. 

In the Sahara all rock is covered with a layer which is always darker than 
the rock itself; this also applies in the case of the different kinds of sand- 
stone, the natural colour of which is white. This forms a protective layer 
and is called desert patina. According to geologists it could only be formed 
in a hot and humid climate, its development being due to an amassing of 
oxides: iron oxide, manganese dioxide, and various carbon dioxides, 
which dissolve in the rain-water that soaks into the rock and are brought 
back to the surface by capillary action, insolation and evaporation. If 
one chips a rock in the Sahara, the cut will have the original light colour 
of the rock, while the adjacent parts of the surface will be darker; from 
this it follows that the more recent an engraving the lighter the patina, 
and vice versa. Thus on a rock covered with engravings it is possible to 



106 




Elephant. Engraving with polished lines. The elaborate ear, with the folds shown, stands out as 
though in relief. Bubalus or pastoralist period. Bardai. Height approx. y—8 in. 



107 



EXTINCT 
ANIMAL SPECIES 



distinguish with certainty between the early works and those of a later 
date, and to arrange them all chronologically. But the reservation must 
be made that this method is not applicable in every instance, for there 
are various factors that can lead to erroneous conclusions. For patina does 
not always form evenly. First the position of the rock in relation to the 
sun has to be taken into account, since those parts exposed to its rays 
are more liable to oxidization. Secondly, patina is formed with varying 
rapidity and produces different shades of colour according to the geolog- 
ical structure of the rock and its oxide content. Thus on many engravings 
depicting camels the patina is as dark as on others representing elephants; 
but as we know that the camel was introduced into the Sahara at the 
beginning of the Christian era, the engravings can only be relatively 
recent, whereas those of elephants are Neolithic. This method must 
therefore be applied with caution where we are not dealing with engrav- 
ings on the same rock or stone slab. Nevertheless, examination of the 
patina, supplemented and controlled by other methods, is an invaluable 
means of establishing the age of a work. 

Further guidance is provided by the species of animals represented. Here 
the criterion is whether they are still to be found in the Sahara or not. 
The earliest discoverers of animal pictures in southern Oran and the 
Sahara were struck by figures of extinct species such as elephants, rhinoc- 
eroses, cattle, etc., and concluded that these were early engravings which 
testified to a subsequent change in the climate. 

The explorer Gerhard Rohlfs, however, who was well acquainted with 
the western Sahara and saw a number of the engravings, attached no 
significance to these figures of elephants, cattle, etc., for, as he said, "these 
pictures are in just the same category as the primitive reproductions of 
steamships that I have seen at Tafilalet and Tuat; they were drawn by 
pilgrims who had returned from Mecca to give their fellow-countrymen 
an idea what a ship looked like." 

Gerhard Rohlfs has had followers, some of whom hold that the engravings 
are the work of travellers who had seen elephants and rhinoceroses in 




Fig. 57 — Stone sculpture. Recumbent 
ox. Tarxerouc, Hoesiar 



108 



the Sudan or chariots in the Roman towns of North Africa. This view 
is superficial, to say the least, and will not be shared by anyone who 
has seen the'Saharan engravings and studied them in situ. It will not stand 
up to serious examination, especially since it takes no account of the large 
number of rock pictures of this kind that have now been recorded, which 
if this theory were true would presuppose an equally large number of 
travellers with the talent of relating their adventures in artistic form; 
it also overlooks the fact that prehistoric tools have been found close to 
the sites of many engravings. 

In actual fact the species of fauna portrayed in the rock pictures is of 
the greatest importance, in that it enables us to identify the major epochs 
in Saharan engraving and painting. In southern Oran, the Tassili and the 
Fezzan, for example, there is one species of animal that is represented 
particularly frequently. This is the prehistoric or Cape buffalo (bubalus 
antiquus), a species now extinct but known from fossil finds. An engrav- 
ing of a bubalus is in itself testimony of antiquity and cannot be inter- 
preted as it has been by Rohlfs. 

Together with this prehistoric buffalo, executed in the same technique 
and covered with the same patina, we can see elephants, rhinoceroses, 
large oxen with stout horns, wild asses, lions, panthers, ostriches and 
rams, which may perhaps be domesticated. 

In the Wadi Djerat in the Tassili we come across almost the same species 
of fauna, but rams are entirely absent and cattle are depicted very 
frequently. 

As evidence of a later period we find many renderings of domesticated 
oxen, very different as regards technique, patina and style. The elephant, 
rhinoceros, ostrich and lion are still represented, but the prehistoric 
buffalo has disappeared. In comparison with the engravings of the pre- 
ceding period there is a marked and unmistakable change: the great 
wealth of cattle is particularly striking, and can be regarded as evidence 
that these animals were reared on a large scale. 

Later still the domesticated horse appears, at first as a draught animal 
before a cart and then mounted. The ox, where it does occur, is no longer 
prominent; the elephant is only rarely represented, and the rhinoceros, 
hippopotamus and equine antelope have disappeared altogether. And 
when the camel finally appears on the scene we only find it in the com- 
pany of species now found in the Sahara (or in adjoining regions inhabited 
by nomadic peoples), such as the moufflon, addax, oryx, gazelle and 
ostrich. The absence of all pachydermata is proof that the climate had change of climate 
changed and that the Sahara now became a desert. 

109 



Alongside figures of animals we also find human beings, whose weapons 

and armour help to determine the date of the pictures. 

In the engravings belonging to the bubalus period humans are depicted 

armed with boomerangs, axes — the Neolithic form of which can clearly 

be seen — and (less frequently) bows. 

In the period when cattle were represented the chief weapon was the 

bow, whereas boomerangs and javelins occur only here and there. 

With the introduction of the domesticated horse we see men armed solely 

with lances and round shields, which survive until after the appearance 

of the camel — although round shields only disappear gradually, making 

way for swords and fire-arms. 

Ox. Probably a domesticated animal, for it has an ornament round its neck. The human figure 
standing before it is holding a curved object, possibly a palm-branch. The lines of the engraving 
are polished smooth. The superposed engravings are stippled; the figure of which only part is 
depicted here represents a small camel from a later period. Bubalus period. Wadi Djerat, Tassili. 

Length of the animal 5 ft. 3 in. 




In conclusion it must be mentioned that the style varies greatly from 
one stage to the next. As has already been pointed out, the older engrav- 
ings are technically and artistically more accomplished, and we can clearly 
trace a steady decline in rock art. 

The oldest engravings in which the bubalus, elephant, rhinoceros and Xatwalistic style 
hippopotamus are represented are remarkably naturalistic; the elaborate 
care given to detail reflects a highly-developed sense of observation and 
admirable knowledge of the animal depicted. No doubt hunting was the 
main occupation of those who were able to produce engravings of this 
quality; the chase, the ambush, and the setting of traps heightened their 
sense of observation, as it did with the rock painters in the Franco- 
Cantabrian caves. 

The engravings of this period are not equal in quality. Some of them 
are the work of great artists, but others are by beginners or men of lesser 
talents. In addition to naturalistic figures there are others schematic- 
ally drawn which may have been produced by genuine artists with crea- 
tive imagination or by incapable dilettantes. There is an exception to 
every rule, and in all the schools of art in the Sahara and in every period 
we find works which are of differing styles. 

In each stage of stylistic development there is a distinct tendency to 
represent objects in certain dimensions. Thus in the oldest engravings 
animals and humans are frequently rendered life-size, sometimes even 
larger than life. In the Wadi Djerat there are rhinoceroses as much 
as 26 ft. 6 in. long, humans over 11 ft. tall, and a group of giraffes of 
which the tallest measures 23 ft. and covers an area of 98 square yards. 
But in this case, too, there is no hard and fast rule, and within the same 
group one may find, in the Wadi Djerat as well as in southern Oran, small- 
scale works with dimensions varying between 1 ft. 6 in. and 3 ft. 

There are different reasons for these variations in size. They are derived Variations in size 
either from the artist's individual conception or his inability to render 
the subject on a large scale; another reason may be the gradual decline 
in the rock art of these regions. Undoubtedly these engravings were 
produced during the course of countless centuries, and each of the periods 
which has the same style and subject-matter may have lasted for several 
millennia — so that even a similarity in the patina does not permit conclu- 
sions to be drawn with regard to chronological sequence. 
In the group of figures of domesticated cattle the style appears less 
naturalistic than in the preceding period, and the attention to detail is 
less marked. Not infrequently one finds legs without hooves, or feet that 
are incomplete. One may observe a certain stiffness in the rendering of 




Heads of pelicans. Birds are depicted relatively seldom in the rock art of the Sahara. Bubalus 

period. Wadi Djerat, Tassili. Height i ft. 7% in. 



form, a certain conventionalism reminiscent of children's drawings; thus, 
for example, ears are placed above the body, making them look like the 
Twisted perspective wings of butterflies, or the horns of oxen are rendered in twisted perspec- 
tive, i.e. full face, whereas the body is drawn in profile. In general they 
are represented on a smaller scale than in the preceding stage, and range 
from 2 ft. to 3 ft. 6 in.; but these are average measurements, and both 
larger and smaller figures can be found. 

With the appearance of the horse a new style develops, related to a special 
technique, making it easy to distinguish this group of engravings from 
the others. Many figures are represented in a naturalistic style, which is 



Fig. 38 — Figures in double triangle style, typical of Ti- 
besti. Rock engravings. Aozon, Tibesti. After Th. Monod 



dot inferior in quality but manifests a certain uniformity in the case 
if horses rendered 'at the flying gallop'. These pictures are on a smaller 
scale, with the sides generally measuring between 10 and 20 in. in length; 
there are also some exceptions, and human beings in particular are 
xxasionally more than 3 ft. tall. The figures have no contour lines but 
:he entire surface is pecked and frequently polished as well. 
IVhen chariots are replaced by mounted horsemen the style forfeits some 
)f its quality and the human body is rendered in a most conventional 
nanner; by two isosceles triangles placed one upon the other and joined 
>y their vertices. This has been called the double triangle style; either 
he entire surface is pecked and polished, or alternatively the lines only 
ire pecked out. 

rhe appearance of the camel seems to mark the beginning of a period of 
lecadence in engraving; if some of the pictures are at first not devoid 
}f a certain quality, most of them manifest a rapid decline towards 
ichematization and come to resemble children's drawings, 
rhis summary of the various factors that make it possible to classify the 
mgravings into various groups applies particularly to the central Sahara, 
/vhere almost all types of engraving can be found; but there are also 
ocal styles and techniques which, though still coming within the general 
xamework, deserve special consideration. 

rhe first attempt at classification was made by the geologist G. B. M. 
Flamand, who studied the engravings of southern Oran and published 
1 standard work on them which is still today fundamental for any 
lystematic treatment of this most complex subject. 3 
rhe ethnologist Frobenius also studied the question of the engravings 
in southern Oran and the Fezzan, but his works opened up no fresh 
ipproaches as regards the classification or chronological sequence of the 
/arious stages. 4 The interpretation which he gave, in terms of the psychol- 
)gy of the different peoples, was greatly weakened by the fact that he 
'ailed to appreciate the differences between the groups of engravings; 
lis disregard for their relative antiquity deprived his final conclusions 
if all real value; he was, as it were, attempting to juggle with two packs of 
cards at once. 




"3 



A more reliable work was written by Professor Th. Monod (of the 
Museum of Natural History in Paris), who spent 18 months at the Adrar 
Ahnet, a spur of the Hoggar. He made an inventory of the engravings 
in this region and as a result of his studies evolved a classification which, 
CLASSIFICATION like that of Flamand, was based upon the criteria set out in the preceding 
pages. 

The work of these experts, supplemented by later studies, made it possible 
to define four different stages: 

I. Bubalus or Hunter Period 
II. Pastoralist Period 

III. Period of Warriors with Chariots and Horsemen, 
or Horse Period 

IV. Camel Period 

Crouching lion, apparently lying in wait, about to spring. The patina of the lines is not so dark 
as in earlier engravings. It may date from an intermediary epoch between the bubalus and 

pastoralist periods. Wadi Djerat, Tassili. Length 1 ft. g3/ 4 in. 









114 



The most important characteristics of these stages may now be outlined. 

1. Bubalus or Hunter Period. The presence of the bubalus in the 
engravings of this stage is an essential criterion, since the bubalus belongs 
to a species that is now extinct. The designation 'hunter period' implies 
that hunting formed the main means of livelihood during this period, 
since all the animals represented (elephants, rhinoceroses, hippopot- 
amuses, giraffes, large antelopes and ostriches) belong to the category of 
wild fauna. 

The ram is also represented in rock pictures of this period, but it is not 
yet possible to say conclusively whether this is a wild species or one that 
had been domesticated. In southern Oran figures of oxen have now been 
discovered as well; in the engravings in the Wadi Djerat this animal 
appears simultaneously with elephants, rhinoceroses, etc., although there 
are many indications that it was already at this time a domestic animal. 
It might perhaps be preferable in future to call this merely 'the bubalus 
period' in order to avoid the confusion and generalizations that result 
from a classification based upon the characteristic feature of human 
activity in a given period. 

Apart from the fauna represented, considered in detail above, the main 
distinguishing characteristics of this period are: 1. naturalistic style; 2. 
contour formed either by a polished groove with a V-shaped or levelled-off 
U-shaped cross-section, or alternatively by regular pecking out of holes 
not connected with one another; 3. dark patina; 4. large scale, but found 
together with engravings of average size; 5. men shown armed with clubs, 
boomerangs, axes or bows, but never with javelins. 

II. Pastoralist Period. 1. The fauna are the same as in the preceding 
period, with the exception of the bubalus, which is no longer to be found; 

2. the style becomes semi-naturalistic, but one also comes across schematic 
figures, probably of later date; 3. the contour is represented either by 
polished lines with a levelled-off U-shaped cross-section (rarely V-shaped), 
or by regular pecking out of disconnected holes; the technique, however, 
is inferior to that of the preceding period, and the polished grooves are 
neither so deep nor so regular; 4. the patina is dark, generally a little 
lighter than that of the rock, but sometimes of identical colour; 5. the 
engravings are of average size and (with a few exceptions) vary between 
1 ft. 6 in. and 4 ft. in length; 6. the human figures are armed with bows. 

III. Horse Period. This is divided into three sub-periods: 

A. Chariot sub-period. 1. The large pachydermata have disappeared — 
apart from the elephant, which is only occasionally depicted; domesticated 
cattle still appear; moufflons and tame dogs occur very frequently. The 



Bubalus or 
hunter period 



Pastoralist period 



Horse period 



"5 



equine antelope has likewise disappeared. 2. The style, at first still semi- 
naturalistic, becomes conventionalized. The older chariots are exquisitely 
drawn and have only one shaft; the chariot horses are almost always depicted 
in profile, but some are viewed from above with the animals standing back 
to back. Several chariots drawn by oxen have also been recorded. In the 
more recent engravings chariots are represented schematically, very often 
Fig. 46 only by the wheels and shaft; some have several shafts and may have 
served useful non-military purposes. The men represented together with 
these chariots are schematized in the double triangle figure already men- 
tioned. 3. As far as technique is concerned, the whole surface is pecked, 
with the holes close together and for the most part overlapping one 
another, and then polished. 4. The patina has a dark chamois colour. 

5. The engravings are small-sized, the sides varying in length between 10 
and 20 in.; but there are also some human figures measuring more than 
3 ft. in height. 6. The weapons are very different from those in the preced- 
ing periods: javelins and round shields appear, but there are still some 
bows to be seen. In combination with chariots of later date men are shown 
with a knife dangling from their forearm, which in shape resembles the 
dagger carried by the Tuareg. 

B. Horseman sub-period. The chariot is superseded by the mounted rider, 
although there are still a few schematically drawn chariots in this period. 
1 . There is no noticeable change in the fauna. 2. The style is semi-naturalis- 
tic in the case of the animals but schematic in the case of the human 
beings, who are represented in the form of double triangles. 3. The 
technique consists of fine pecking, with the holes usually overlapping one 
another, but less elaborately executed than in the chariot sub-period; the 
inner surface is rarely polished. 4. The patina is chamois-coloured. 5. The 
engravings are small-sized, varying between 10 and 20 in. But there are 
also local schools where the engravings measure as much as 3 ft. or more 
and the style is almost schematic — as, for instance, at Tit in the Hoggar. 

6. The men are armed with a javelin, round shield and a knife dangling 
from their forearm; but several bows have also been found. The type of 
weapon, moreover, varies from area to area, and three zones can be 

Plate p. 126 clearly distinguished: Mauritania, the central Sahara and the Tibesti area. 

Plumes, which previously appeared here and there together with chariots, 

now form a decorative head-dress worn by all warriors. In the central 

Sahara Libyco-Berber written characters now also appear; these were 

Fig. 36 introduced by the horsemen. 

C. Horse and camel sub-period. This is the period in which the camel 
appears, but the horse is still in use as well. 

116 




^ '• : *0'i§N» 



Ox with thick horns pointing forwards. The lines are very deeply incised. The style is mediocre. 
Late pastoralist period. The smaller figure superposed upon the larger one dates from the same 
period. Bardai. Height from foreleg to tip of horns approx. 2 ft. 3 in. 



117 






i. There are no changes as regards the species of fauna represented, but 
figures of cattle become ever rarer. 2. The semi-naturalistic style becomes 
decadent, but a few fine figures are still to be found. 3. The technique 
employed consists of fine pecking (which is, however, cruder than in the 
preceding period), frequently covering the entire inner surface of the 
figure; the surface is rarely polished. 4. The patina is chamois-coloured. 
5. The engravings are small, varying between 7 and 16 in. 6. The weapons 
are the same as before, and the warriors still wear plumed head- 
dress. 

Camel period IV. Camel Period. This is the most recent period, which is still in 
progress, for the present-day population of the Sahara continues to 
practise the arts of rock engraving and painting. 

1. The fauna comprise only those animals still extant in the Sahara and 
adjacent areas to the south: the antelope, oryx, addax, gazelle, moufflon, 
ostrich, zebu and goat; the horse occurs rarely (in Mauritania, however, 
more frequently than elsewhere) and is now depicted with a stirrup 
saddle of the Arab type. Figures of giraffes are still to be found in the 
southern districts of the Air and the Adrar des Iforas, where this animal 
was extant until only a few years ago. 2. The schematic style is on a par 
with that of children's drawings; human figures in the form of double 
triangles become rare and are superseded by linear forms. 3. The 
technique consists in roughly pecking out the entire figure. 4. The patina 
is very light, almost white. 5. The engravings are on a very small scale, 
varying between about 6 and 8 in. 6. In the older engravings of this 
period almost the only weapon found is the javelin, which is still in 
use today, but here and there one also comes across swords, such as are 
worn by the Tuareg, as well as fire-arms. 

PAINTINGS The classification given above was worked out on the basis of a study of 
engravings, before the discovery of the numerous rock paintings; if these 
had been taken into account this would doubtless have influenced the 
general principles of our theory. 

Rock paintings are not found everywhere. In contrast to engravings, 
which occur wherever there are rocks capable of taking them, paintings 
are only located in places protected from inclement weather, owing to 
the perishable nature of the materials used. As in all other parts of the 
world, the sites containing rock paintings are either in caves or beneath 
overhanging rocks. In the Sahara there are no real caves, but only 
overhanging or hollowed-out rocks, which are found solely in areas with 
a specific mineralogical structure, especially in sandstone massifs. Some 
paintings have also been identified in the granite of the Hoggar, but 



118 



this is exceptional, whereas they occur in thousands in the Silurian and 
Ordovician sandstone formations in the Tassili-n-Ajjer. 
The first rock paintings of any importance were discovered on the Ennedi 
plateau and the Djebel Ouenat, i.e. south and east of the Tibesti massif. 
These were soon followed by further discoveries, at first in the Tassili, 
then in the Hoggar, and finally in Mauritania and the Tibesti. Most of 
these paintings could be classified according to the major categories of 
rock art as established on the basis of the engravings; judging by the 
species of animals portrayed, they fell into the cattle period, chariot and 
horseman period, and camel period. But in 1956 and 1957 new discoveries 
were made in the Tassili that surpassed in number, quality and variety 
everything found hitherto. During an expedition lasting 16 months more 
than 10,000 pictures were registered and photographed to scale in their 
natural colours. On a number of these frescoes one may identify several 
strata or superposed paintings. One of them has no less than 16 such 
layers and makes it possible to establish the chronological sequence of 
the various styles and periods. The analysis of this important find is still 
in progress and far from complete, but we are already able to distin- 
guish between some 30 different styles, i.e. far more than have hitherto 
been identified in the engravings. More than half of them can be classified 
into the three major epochs of cattle, horses and camels, but the rest 
cannot simply be attributed to the so-called bubahis period. Many of the 
paintings have numerous features in common with this period and may 
belong to it; but the majority of them seem to originate from a long 
intermediate epoch between the bubalus and pastoralist periods. From a 
stylistic point of view these paintings display characteristics of negro art 
(symbolism and highly-developed animism); and 
several representations of typical negro masks lend 
support to the view that these paintings are the 
work of negroes. 



Fig. 59 — Rock engraving. Aouilalam, Adrar des Iforas 




U9 




Ox with long horns. Wide, deeply incised lines. Schematic decadent style. Late pastoralist period. 

Bardai. Height from hoof to dorsal line approx. 2 ft. 9 in. 



120 



In this most unusual complex of rock paintings, in which the human 
face is uniformly portrayed in a symbolist style,- round and featureless 
(resembling somewhat the hooded inhabitants of Mars in Jules Verne's 
tale!) and later takes on the features of animistic masks, it seems as though 
we are confronted with the earliest works of negro art — indeed, one is 
tempted to say, with its origin. We thus have completely new elements 
that are later to lead to a modification of the present 4-stage classification 
of the Saharan rock engravings and paintings. 

The discovery of the rock art of the Sahara, with its figures of animal 
species that have since vanished from this region, has shown conclusively 
that this area, today the largest desert on the earth's surface, was once 
inhabited, and that in the course of its long history it has afforded shelter 
to peoples having the most varied modes of existence. 
The presence of prehistoric man is, moreover, confirmed by numerous 
finds of Stone Age implements and other traces of human habitation, 
particularly kitchen refuse, in which one often discovers bones of extinct 
animals depicted in the rock pictures. With the aid of these implements, 
which are scattered throughout the area, we are now in a position to 
give an outline sketch of prehistoric civilization in the Sahara. 
We are familiar with the history of the gradual development of Stone 
Age implements in Europe; we know that in the beginning these were 
crudely worked pebble tools or flake tools which later displayed increasing 
refinement as new knowledge was acquired, new needs developed, 
techniques were perfected, and the general conditions of life improved. 
Axes, scrapers and blades, which at first were produced merely from 
roughly hewn flakes, gradually became objects of more elaborate and 
accurate workmanship, and simultaneously decreased in weight. The 
final products of this experimentation were the polished axes and smooth 
flint arrow-heads that were later slavishly copied by the incipient metal in- 
dustry. The first group of these tools is Palaeolithic, the second Neolithic. 
This chronological division, introduced with Europe in mind, eventually 
had to be adopted in other parts of the world, and particularly in Africa, 
for it has been ascertained that in this continent industry developed on the 
same lines, and the same types of implement were produced, as in Europe 
— some small details excepted. Each of these two regions is in fact 
an extension of the other, and even today we do not know in which of 
them the human race first appeared — although Africa has in recent 
years furnished evidence suggesting that it was one of the cradles of 
mankind. However that may be, in the past the Sahara could boast of a 
population as large as, if not larger than, any other region in the eastern 



THE SAHARA 
IN PREHISTORY 



Tools 



121 



Density of population hemisphere. Its density was greatest in the Lower Palaeolithic, at the time 
of the Chelles-Acheul industries; the typical products of that era, heavy 
pieces crudely chipped on both sides and referred to as hand axes, are 
found in all areas: in the ergs (deserts of sand dunes), regs (deserts strewn 
with pebbles), at the outlets of valleys in the mountain massifs, and in 
the coastal areas. These objects are for the most part widely scattered 
about, but in isolated cases they were also found at hearths together 
with remains of bones, among them those of prehistoric elephants, large 
hippopotamuses, prehistoric buffaloes, a zebra that cannot definitely be 
identified, and several fish of the Silurian period, as well as species that 
are now largely extinct. 

The much older and far more primitive pebble-tool industry exists in 
the Sahara as well as in South Africa, where it has been discovered in 
recent years; it represents the first fumbling effort by man to produce 
implements. 

In Africa, as on our own continent, the people of this epoch lived from 
hunting and fishing, and probably also gathered certain kinds of red 
berries and herbs. 
Race What race did they belong to? The question still remains open, for no 
Palaeolithic human skeleton has as yet been found together with im- 
plements from this period. But they probably belonged to the race that 
inhabited North Africa and which we have recently come to call the 
race of Atlanthropos of Palikao. 

The Levalloisian-Mousterian industries, which in Europe were superseded 
by those of the Chelles-Acheul culture, were likewise indigenous to the 
Sahara, but were more localized, particularly in the northern region; they 
have, however, left fewer traces, an indication that the area was now less 
densely populated than before. The Upper Palaeolithic, which cor- 
responds to the Ice Age and reindeer hunter culture in Europe, is re- 
presented in North Africa by clearly identifiable local industries, but these 
do not appear in the Sahara. There is a gap here, the causes of which are 
still unknown; but it seems that the conditions of life, which had been 
so favourable in the Lower Palaeolithic, deteriorated in the subsequent 
period, as is indicated by the low density of population in the Mousterian. 
Change of climate Did the Sahara go through a phase of desiccation on some previous occa- 
sion? This is not impossible, as seems to be suggested by the existence of 
fossil dunes, ascribed by geologists to the final stages of the Palaeolithic. 
Suddenly the scene changes completely! Once again masses of people 
stream into the whole region between the Atlas range and the Niger; 
mountains and valleys are resettled by numerous tribes. This is the 




Giraffes. Wide, crudely incised lines. Mediocre style. Late pastoralist period. El Greiribat. Height 
of taller giraffe approx. i ft. 10 in. 



123 




Fig. 40 — Stone sculpture. Recum- 
bent wether. Wadi Amazzar, Tassili 



beginning of the Neolithic epoch with its polished axes, flint arrow-heads 
and various types of potsherds. Thousands of finds in all regions of the 
Sahara bear witness to this period; they are of the most assorted shapes 
and sizes, indicating that numerous waves of migration took place and 
that the modes of life of the various tribes were very different. The valleys 
were, of course, the most densely populated parts, for they were at the 
same time the trade routes. Ancient settlements of fishermen have 
been identified in the south of the Hoggar, in the Tanezrouft and Tenere; 
here vast quantities of fish-bones and mussel-shells were found. As is 
still the case today on the banks of the Niger, the menfolk fished in the 
deep water and hunted hippopotamuses, whilst at low water the women- 
folk gathered mussels. 

Several skeletons recovered from strata of this period show that the 
fishermen of the Sahara were negroes. In other areas axes with an arched 
blade at right angles to the handle seem to have served as hoes, suggesting 
Agriculture that the soil was tilled on the low-lying land and areas subject to flooding 
by the rivers. The rivers, incidentally, seem to have been deep, at least 
during several months of the year, for otherwise they could not have 
held such an abundance of fish — nor, in particular, of hippopotamuses 
and crocodiles. Remains of these animals are frequently to be found in 
the kitchen refuse left behind by humans on terraces situated high above 
the floor of the valley, which apparently served them as places of refuge 
when the water began to rise. Herbs began to play an ever greater part 
in the diet; this is evidenced by the abundant finds of grindstones, 
crucibles and pestles made of stone, as well as the large number of 
potsherds. 
Hunting Hunting, too, was practised on a large scale, with the bow as the most 
common weapon. Arrow-heads are to be found in the ground everywhere: 
in the valleys and ergs (deserts of sand dunes) as well as in the mountains, 
where the hunters had a wealth of game available. 
Cattle-raising Cattle-raising appears to have been held in high esteem by some tribes; 



124 



although excavations have until recently been unable to provide us with 

much information on this subject, some light is thrown upon it by the 

evidence of the rock pictures. 

It is possible to follow with a fair amount of accuracy the receding of 

the water in the rivers, i.e. the beginning of desiccation. One can note 

the formation of numerous lakes, indicating that the water could no 

longer discharge and was held back in the low-lying land. Furthermore, 

there are traces of settlement on the banks of the rivers, indicating 

that the people were no longer forced to take refuge from the floods 

on the terraces, but descended into the valley to be closer to the receding 

water. In deposits of the Late Neolithic aquatic fauna are completely 

absent. Gradually the desert began to spread, and man was compelled 

to withdraw either to the humid lowlands, where oases were formed, 

or into the mountainous regions, where there were, and still are today, 

springs and water-holes — natural basins that are fairly plentifully and 

regularly filled and refilled with rain-water. 

From our study of prehistoric deposits we can attempt to determine the AGE OF SAHARAN 

c u ■ j • • r .u c u ROCK. PICTURES 

age of the engravings and paintings 01 the Sahara. 

Many authors, influenced partly by the work done on the Spanish paint- 
ings and partly by the extinct species of animals depicted in the old 
engravings, were at first inclined to ascribe the large-sized engravings 
of the bubalus period in southern Oran to the Upper Palaeolithic. It is 
particularly difficult to determine the age of the rock pictures since, in 
contrast to the finds of implements, they do not occur in archaeological 
strata. In Europe, in the Franco-Cantabrian area, sacral or profane im- 
plements, as well as plaques of ivory or limestone bearing engravings, 
were found in archaeological strata which corresponded to the representa- 
tions on the walls of the caves; and by comparing them it was possible 
to establish the antiquity of the rock pictures. But the archaeologist in 
the Sahara has nothing of this sort to assist him. 

Implements discovered in the proximity of the sites can help to determine 
the age of the pictures, although it is impossible to prove that the man 
who produced an implement discovered at the foot of an engraved or 
painted rock must also have been the creator of the work concerned. 
At the sites in southern Oran, which could be examined most thoroughly 
owing to their ease of access, most of the tools discovered at the foot of 
the murals were Neolithic, and no older pieces were found, even in the 
general proximity of the works. In the light of this fact the hypothesis 
that these engravings belonged to the Palaeolithic had to be abandoned, 
and instead they were ascribed to the Neolithic. These were the conclusions 

125 




Archer. The inner surface of the figure is chipped. The three feathers on the head resemble those 
on Libyan warriors in Egyptian figures. The clothing consists of a bell-shaped tunic, fastened at 
the waist and hanging half-way down the thigh. The bow was very rare in this period. Horse 

period. Wadi Djerat, Tassili. Height i ft. 5 1 /3 in. 



126 



reached after several months of study on the spot by Professor Vaufrey 
of the Anthropo-Palaeontological Institute in Paris. 
But Abbe Breuil, the internationally acknowledged authority on pre- 
historic art, was inclined to place them at a slightly earlier date, although 
he did not support the thesis that these works should be ascribed to the 
Palaeolithic; instead, he put their origin at a time of transition from 
the Palaeolithic to the Neolithic. 

The observations of Professor Vaufrey are incontestably correct in so 
far as they concern the general occurrence of Neolithic implements, but 
the complete absence of pottery and the rarity of arrow-heads and polished 
axes suggest a very early Neolithic. 

This question again became acute recently, after a fresh examination of 
the engravings in the Tassili, particularly of the famous sites in the Wadi Sites in the 
Djerat, where more than 4000 engravings were registered, originating ! Tat 

from different epochs and distributed over a length of more than 18 
miles. With each complex of engravings of the bubahis period situated 
near a terrace, crudely worked flakes were found bearing an obvious 
affinity to those of the Levalloisian-Mousterian. There are unfortunately 
only a few typical specimens which, judging from a cursory examination, 
may belong to a very primitive Neolithic; but it is still too early to draw 
final conclusions. In any case, these ' knocked off' flakes, even if they 
originate from the Palaeolithic, show that these engravings are pre- 
historic; they can be ascribed at least to a very early phase of the Neolithic 
following closely upon the Palaeolithic; here one may agree with the view 
of Abbe Breuil. 

It should be mentioned in this connection that the Levalloisian- 
Mousterian industries, which in Europe belong to the Middle Palaeolithic, 
were able to survive in the Sahara up to the beginning of the Neolithic. If 
pieces of the crude tool used in this industry have been found below 
engravings of the bubalus period, and identified as Mousterian in origin, 
this would not make the engravings older but would indicate that those 
who produced them were Palaeolithic men; from an anthropological 
point of view they would thus be distinct from the Neolithic peoples 
who suddenly poured into the Sahara, about whose movements we still 
know so little. 

If one takes as a basis the chronology that applies in Egypt and North 
Africa, early Neolithic would date back to approx. 7000 B.C.; the engrav- 
ings of the bubalus period would thus have originated at this time — 
with a margin of 2000 years either way, since this period must have been 
very lengthy. 

127 



The paintings in 'round-headed style' seem to follow immediately after 
Plate p. 137 the bubalus engravings, if indeed they are not contemporaneous with 
them. Here the student is confronted with the same difficulties as before. 
In view of the fact that the walls on which they appear frequently also 
bear more recent figures of cattle and horses, it is impossible to ascribe 
the tools found at the base of the rock to the creators of these works. 
In any case, wherever a site is stylistically pure and free from alien 
influence, the tools found on the ground nearby are crude axes formed 
by large flakes, the workmanship of which is reminiscent of those found 
in the Campignian culture in Europe. Never do we find any pottery, 
grindstones or pestles, whereas these items occur very frequently with 
the paintings of the cattle period. They may thus be considered early 
Neolithic, although the reservation must be made that it is not yet 
certain whether those who painted the pictures in round-headed style also 
produced these tools. 
Pastoralist civilization We are better informed about the pastoralist period, for tools and house- 
hold implements are to be found in great abundance in the rock-shelters 
which contain their remarkable works of art. Below the overhanging rocks 
where they dwelt the ground is strewn with potsherds, grindstones, 
crucibles, pestles, polished stone axes, and flint arrow-heads. Quite 
frequently remains of ashes are found; these are not always located at 
the same spot, which suggests that the sites were not permanently inhab- 
ited — as is only to be expected with nomadic pastoralists constantly in 
search of new pasture-grounds. This refuse, which at times is as much as 
2 ft. thick, has yielded ample material for study, particularly pieces of 
bone, which are no doubt remains of meals; they are mainly bones of 
cattle, but there are also some from goats and sheep, for these two animal 
species were likewise domesticated. The bones of cattle found in great 
quantities are, so to speak, the visiting cards left behind them by these 
pastoralists which make it possible to identify other debris. The finds 
buried under the ashes are mainly potsherds, arrow-heads, grindstones, 
crucibles and axes similar to those found on the surface, but there are 
also pendants of schist, armlets of the same material, ostrich egg-shell 
necklaces, awls, hollowed-out bones engraved with geometric designs, 
and furthermore a number of small accurately-shaped round discs of 
sandstone; the use to which these latter were put remains a mystery, but 
they may have served as amulets worn by animals. We are thus well 
informed about the material aspect of this pastoral civilization. Remains 
of charcoal have also been found in the deposits and have been subjected 
to radio-carbon analysis. Two tests made on samples taken from different 



128 



sites showed the dates of origin as 3550 B.C. and 2450 B.C. Systematic 
excavation arid more tests would probably yield dates even further apart. 
The pastoraHst period presumably lasted for several millennia. The date 
of its commencement can be put at 4000 B.C., and its termination may 
coincide with the appearance of horsemen around 1200 B.C. 
For the chariot period we have other sources of information: firstly, the 
chronicles of the Egyptian architectural monuments, and secondly 
writings of authors of antiquity, such as Herodotus and Strabo. We are 
now no longer in the prehistoric era but at the beginning of recorded 
history. 

The domesticated horse was only introduced into Africa at a very late 
stage, by the Hyksos on their conquest of Egypt, about 1550 B.C. But 
the chariots drawn on rock walls in the Sahara are more recent and 
originate from the time of a second invasion, that of the 'people of the 
sea'. These came from Crete to Cyrenaica, allied themselves with the 
Libyans settled in that area, and launched an attack upon Egypt. This 
makes it possible to interpret this style, in which horses and other animals 
are rendered 'at the flying gallop', the human figures have exaggerated 
slender bodies, and are armed with javelins and round shields: all these 
elements also occur in Cretan or Mycenaean art. The association of this 
artistic style, and the weapons represented in it, with the invasion of the 
'people of the sea' lends this hypothesis a very high degree of probability. 
It should, however, not be inferred that Cretans were also responsible for 
these pictures, although we know that the invaders, who landed with 
their families and their chariots, were driven back and compelled to 
withdraw into the territory of their Libyan allies. The influx of popula- 
tion into such a narrow strip of land as Cyrenaica probably led to 
emigration to more southerly areas which were less densely populated 
(and if anything wealthier), and to interbreeding between different tribes. 
It was through contact with the 'people of the sea' that the Libyans 
acquired the most characteristic features of their art; thus the rocks of 
the Tassili attest to the sudden efflorescence of a new school devoid of any 



Chariot period 



Invasion by the 
'people of the sea' 



Fig. 41 — War chariot. A fine 
illustration of 'flying at a gallop' 




129 




Fresco. In the centre two lions devouring an ox; a jackal appears to be waiting for the remain: 
of the lions' meal. There are also other figures of cattle, warriors armed with javelins and rounc 
shields, Libyco-Berber characters (left), a moufflon and several camels of more recent date thai 
the lions and cattle. Horse period to Libyco-Berber period. Wadi Djerat, Tassili. Total length 

of fresco approx. 8 ft. 3 in 



130 



connection with the preceding pastoralist school. Particularly astonishing 

is the wide diffusion of works in this style, which are not only to be found 

in the Tassili and Hoggar, but also cross the borders of the desert region 

of Tanezrouft (where there are five figures of chariots 'at the flying gallop' 

at the spring of Ti-m-M issao), and reach the Adrar des Iforas, and almost Fie 41 

reach the Niger. 

There can be no doubt that the arrival of the 'people of the sea' in 

Cyrenaica had decisive consequences for the history of the Sahara, and 

that the introduction of war chariots gave the Libyans great military 

superiority, enabling them to spread their power rapidly to the middle 

reaches of the Niger. Ten centuries before the birth of Christ war 

chariots crossed the Sahara, and white men who had hitherto occupied 

only the most northerly regions of Africa penetrated into the heart of 

the Dark Continent. 

In the 5th century B.C. the Greek historian Herodotus recorded that the Herodotus 

Sahara was inhabited by Libyans and described the customs of particular 

tribes such as the Nasamones, a tribe from Augila, the Asbystae, Macae, 

Lotophagi and others living in the coastal districts, and the Garamantes, 

Atarantes and Atlantes in the interior of the country. The war chariot 

was familiar to most of these peoples; Herodotus records that the Asbystae 

womenfolk were skilled at driving them and that the Garamantes in the 

Fezzan hunted down the Troglodytes with chariots drawn by four horses. 

The Troglodytes were no doubt the inhabitants of the Tibesti, that is to 

say Teda or Tibbu. The Atarantes and Atlantes lived in the Tassili or 

the Hoggar. The chariot continued to exist for a relatively long time and 

was still in use among the Libyans who served in Hannibal's army. We 

do not know when it disappeared from the Sahara, but it certainly was 

still current in the period of the decline of engravings depicting horses. 

But it should not be concluded that it was still in use at the time when the 

camel came on the scene, for there is no evidence that this is so. 

The riding of horses was apparently not known to the Libyans before 

the time of Herodotus in the 5th century B.C., for he makes no reference 

to this. But the oldest known equestrian figures in the Sahara are in the 

'flying gallop' style, the same style as that of the chariots. The use of 

horses as mounts therefore seems to have already been introduced to the 

Libyans earlier, but not to have met with immediate general acceptance. 

One interesting point of detail may be noted here: these earliest Libyan 

horsemen led beside them a second horse, i.e. a near horse to replace 

their mount if it were killed in battle. 

It is reasonable to assume that the chariot was only superseded by mount- 






1 3 1 




CHRONOLOGICAL 

COINCIDENCE OF 

ENGRAVING AND 

PAINTING 



Two modes 
of expression 



Fig. j2 — Stone sculpture. Sacred stone 
with human head. Ouan-Sidi, eastern Erg 



ed riders after the time of Herodotus. But horses are also represented in 
the Sahara at a later epoch, long after the introduction of the camel into 
that region. Its disappearance was doubtless due to the increasing des- 
iccation. Strabo, who died in 21 A.D., records that the desert peoples 
travelled on horseback but as a precaution took with them a bag of 
water, which they tied under the belly of their steed. 
Opinions are greatly divided with regard to the time when the domes- 
ticated camel was introduced into the Sahara. Some authors give a very 
early date, others a very recent one. The written records appear to support 
the view that this occurred relatively recently: the camel is first mentioned 
in an account of the battle of Tapsus, in which the Romans took from 
King Juba 22 of these animals as booty. From this time onwards the 
camel becomes indigenous throughout North Africa. 
Examination of rock engravings and paintings shows that the camel 
was never represented in the early periods and makes its first appearance 
together with a horse. This strongly supports the argument that it was 
introduced at a late juncture. Today the camel is the only animal capable 
of traversing the Sahara under its own power. Its introduction (appar- 
ently from the Arabian peninsula) and diffusion were doubtless a con- 
sequence of the desiccation of the Sahara. 

All this gives an idea of the chronological framework in which the rock 
engravings and paintings of the Sahara are set. 

The fact that both rock engravings as well as paintings appear in the 
Sahara intimates that both these modes of artistic expression were in use 
simultaneously and that painters and engravers existed at the same time. 
We do indeed know of epochs in which both techniques were employed 
side by side, but this was not always the case. It is worth examining this 
question more closely, for it is of importance if we are to understand 
what kind of peoples lived in the Sahara in early times and the migrations 
that took place in this area. 

From the bubalus period we only know of engravings, and as yet no 
paintings have been discovered that were produced at this time. All we 
can say is that some bubalus engravings are superimposed by paintings — 
as, for example, at Kef Bou-Bekeur near Tiaret in southern Oran. 
Research now being carried out in the Tassili may perhaps eventually 



132 



lead to analogies being established between the paintings in the round- 
headed style and the bubalus engravings, but there is as yet no evidence 
of this. 

The cattle period surpasses all others in wealth of artistic works; the 
engravings and paintings from this period may be numbered in thousands. 
One might almost think that these works all originate from the same 
era and were produced by the same tribes according to their various 
whims and circumstances. In his study Rock Drawings H. A. Winkler 
advanced the thesis in regard to Kargour Thai in the Djebel Ouenat that 
the paintings were the work of women and the engravings of men. But 
this is unlikely, for if that were so paintings and engravings would have 
had to appear simultaneously in one zone, which is far from being the case. 
In this connection an important discovery has been made in the Tassili, 
the area of the Sahara containing the largest complex of paintings of the 
cattle period. In this zone, which boasts of more than 10,000 painted 
figures, there are practically no engravings dating from the same period, 
although rock faces suitable for this purpose are to be found there in 
abundance. From this significant fact it may be inferred that the painters 
and engravers of the cattle period were not necessarily members of the 
same community and that there must have been groups of painters and 
groups of engravers existing independently of one another. 
If there are nevertheless some sites where engravings and paintings are 
to be found side by side, these are rare exceptions. The heterogeneous 
character of these two techniques is evidenced not only by the geographical 
distribution of these works but also by the ideas that inspired them. Thus, 
for example, human beings are frequently portrayed in paintings, but 
rarely appear in engravings. The paintings are always scenic compositions, 
whereas in the case of engravings objects are almost always represented 
singly. 

It is difficult to explain these contrasts except by the existence of two 
groups of people of different origin, one of whom introduced cattle- 
raising to the other; this would suggest that as the pastoralists advanced 
into the central Sahara they came across inhabitants of that area not yet 
familiar with cattle. 

But as far as the chariot, horse and camel periods are concerned it may 
be taken for granted that a chronological correlation existed between 
paintings and engravings. 

The distribution of the various genres of rock pictures must be considered 
with reference to the work of each individual period. 
The bubalus period only boasts of engravings, the majority of which are 



Representation of 
human beings 



DISTRIBUTION OF 
ROCK PICTURES 

Bubalus period 



large, although there are also some of average or even very small size. 
Three centres of art are known in this period: southern Oran, the Tassili 
and the Fezzan. Some engravings from this period may also exist in the 
Hoggar, the Tibesti and in southern Morocco, but no conclusive evidence 
of this is as yet available. However this may be, such a pattern of distribu- 
tion is most curious, for the various centres are separated from one 
another by vast areas in which bubalus artists have left no trace of their 



Warriors, mounted on horses and camels, and armed with javelins and round shields. The horses 
carry saddles with back supports (so-called Arab saddles), an indication that these figures date 
from the period after the hegira (Mohammed's flight from Mecca). Wadi Djerat, Tassili. Maximum 

height of figures 9I/4 in. 




'31 



activity; this implies that they were probably not members of the same 
tribe, even though the techniques applied are the same and the styles 
are akin to each other. This seems to be corroborated by the fact that 
the subjects treated in these engravings are not the same, from which it 
follows that their religious conceptions must also have been different. 
In southern Oran human figures are frequently rendered in an attitude 
of supplication and in combination with animals, such as, for example, 
the bubahis, the ram (shown with a disc between its horns) and the lion; 
these figures apparently had some religious significance. Neither in the 
Tassili nor in the Fezzan is this motif to be encountered. The ram with 
a disc is also absent, whereas the ox with symbols on its horns appears 
quite often; a few exceptions apart, it is not, however, depicted in com- 
bination with human figures. In the Tassili spiral-like ornaments are 
often found in combination with animals. This motif may have had 
some symbolic or religious character; its significance remains a mystery. In 
southern Oran and in the Fezzan, incidentally, spirals are not known. 
In the Tassili, on the other hand, there are a large number of figures 
connected by a line drawn from the genitals of one to the genitals of 
the other, of such a grossly indecent character that one has qualms in 
presenting them to the reader. Scenes in this genre also appear in the 
Fezzan, but are, however, completely absent in southern Oran. All these 
rather marked diversities confirm the fact that only a very slight affin- 
ity existed between the three groups of artists. 

We do not know who produced these fine engravings, and their distribu- 
tion, as we have just seen, is of no assistance in answering this question. 
But the discovery of an engraved plaquette in a Capsian layer in southern 
Tunisia, bearing a figure drawn in a style reminiscent of the figures in 
southern Oran, permits the conclusion that this art originated in the 
Capsian civilization that sprang up on the plateau land of western Algeria 
and Tunisia, among men related in type to Cro-Magnon man. 

Cro-Magnon man belongs to the white race, so that these engravings were white race 
the product of white men. This assumption is by no means so surprising 
as it may appear, for some of the figures portrayed in the engravings of 
southern Oran have a profile which, though imperfectly executed, can 
be identified as European rather than negroid. The same applies to 
the figures of humans in the Tassili, the majority of whom have European 
profiles. Some of them possess negroid features, but this is not necessar- 
ily significant, for many human beings in these pictures are portrayed 
with ape-like animal heads. Yet these animal heads provide no evidence 
of the existence of animistic cults as practised by the modern Bushman. 



135 



'round-headed men 



Nor should one forget the role played by certain animals in Egyptian 
culture. 
Period of The paintings of the so-called 'round-headed men' are later in date than 
the engravings of the bubalus period and are doubtless the work of 
peoples of different origin. This singular school is, so far as we know 
at present, confined to the eastern part of the Tassili-n-Ajjer, i.e. to the 
region of the Sahara in which there are the largest number of overhang- 
ing or hollowed-out rocks. These paintings, though small to start with, 
gradually increase in size and finally attain extraordinarily large dimen- 
sions; with their figures of human beings measuring more than 16 ft. 
in height and of cattle and other animals in natural proportions, 
these are among the largest prehistoric paintings anywhere on earth. 
In order to produce such works sufficiently extensive rock surfaces were 
required such as are only to be found in the Tassili; it may even have been 
the case that the extra-large protected rock walls of the Tassili gave the 
painters the idea of working on a large scale. The absence of similar 
hollowed-out rocks in adjacent areas accounts for the marked localization 
of this school, but we cannot simply conclude from this that the tribes 
to which these artists belonged were not settled in other regions as well. 
These paintings, as already mentioned, constitute the earliest elements 
of negro art hitherto known, which again is evidence of the fact that 
the Sahara, its northern districts included, was once inhabited by coloured 
people. 

This negro art must have lasted for several millennia; during the 
last phase of its development it shows a marked tendency towards or- 
namentation, a feature that can be seen very distinctly in the 'White Lady' 
of Auanrhet. This is one of the finest works and portrays the goddess 
of an agrarian cult, possibly of Isis. At the end of this most advanced 
stage one can find traces of foreign Egyptian influence: the human figures 
are more fluent and elegant in style, and wear head-dress resembling 
that worn by the Egyptians. In the Tassili this unexpected contact between 
negro and Egyptian art has given us such impressive works as the portrait 
of a woman, no doubt a high-ranking personage, whom we have christen- 
ed 'Antinea' and whose pure Greek profile make her seem a strange and 
inexplicable phenomenon in this negro setting. 

Where did these coloured peoples spring from? The fact that this genre of 
painting is confined to the Tassili and is absent from other areas prevents 
us as yet from putting forward hypotheses as to their origin and the 
routes they took on their migrations. All that can be said with certainty 
is that they are related to the negro peoples of present-day West Africa. 



136 




Masked woman, wearing ornaments above the knees, anklets and armlets. She is clad only in a 
girdle, the ends of which hang down to her knees. The curved object that she is holding in her 
hands cannot be identified; it may be a bowl. Period of 'round-headed men'. Sefar, Tassili. Height 
approx. 2 ft. 3 in. 



137 




Fig. 4} — Stone sculpture. Recumbent 
mammal (wether?). Silet, Hoggar 



Pastoralist period 



Plate p. 120 
Plate p 1 17 



Origin of the 
pastoralists 



Works of art originating from this period are spread over a particularly 
wide area. They are to be found in all the mountain massifs, as well as 
in the hilly country west of the Nile, in southern Oran, southern Morocco, 
Rio de Oro and Mauritania. From this it can be inferred that at some 
given point in history the whole Sahara was inhabited by pastoralists 
whose economy was based upon cattle and to a lesser extent upon sheep 
and goats. The cattle was of two species: firstly the bos africanus, an ox 
with long horns; and secondly the bos brachyceros, an ox with stout 
horns. The sheep seem to have belonged to a species with a fat tail, the 
Berber sheep and not the ovis longipes depicted in the engravings of 
southern Oran and still reared today by the shepherds of the Sudanese 
steppe and the Sahara; it was an ovis longipes which incarnated the Libyan 
deity Amon, later adopted by the Egyptians. 

It has been held that cattle were domesticated in the Sahara. This is very 
unlikely, and the most recent research suggests that the wave of pas- 
toralists that streamed into the Sahara came from the adjacent Upper 
Nile region. The paintings appear to be older than the engravings; their 
most important centres of distribution are the Tassili and Tefedest 
massifs, parts of the Hoggar. There are also some in the Ennedi and in 
Mauritania, but these are more recent products, indicating that the 
immigrants did not reach these areas until much later. It may not be 
mere chance that the main centres of painting are situated in the east, 
on the borders of these massifs; from this it can be inferred that the cattle- 
raisers came from the east. At first they must have arrived in the Tibesti, 
then have moved on to the Tassili and Hoggar, and finally have reached 
the Djebel Ouenat. It may also be mentioned that paintings of cattle do 
not occur in the Air and Adrar des Iforas massifs on the border of the 
steppe, and that the engravings discovered in the Adrar des Iforas belong 
to a school centred in the Hoggar. 

These facts refute the hypothesis advanced by Rhotert, a follower of 
Frobenius, according to whom the pastoralists advanced into the interior 
of the Sahara by way of the Sudanese steppe after conquering the Air 



and Adrar des Iforas massifs. In fact they came through the valleys and 
over the massifs of the central Sahara, for in this area there were at that 
time suitable pasture-grounds which made a more southerly route unnec- 
essary. Everything points to the fact that the Sahara was still covered by 
patches of grassland and that the pastoralists painted the animals which 
dwelt around them, such as rhinoceroses, elephants, giraffes and even 
hippopotamuses. The latter appear in pictures representing hunters in 
a boat, indicating that in that epoch the vegetation was very luxuriant 
and the rivers deep enough for these animals to live in them. 
These pastoralists, especially the painters among them, were artists of 
note. Their art, contrary to that of their negro predecessors, is devoid of 
any symbolism or animistic influence; with their astonishingly keen sense 
of observation, they were able to create one of the finest schools of natur- 
alistic art. Figures rendered singly are rare, whereas composite figures, in 
which humans occupy a prominent place, occur frequently. Strict atten- 
tion is paid to perspective, to an extent that is nowhere found in 
Palaeolithic Franco-Cantabrian art; indeed, the pictures often give the 
impression of photographs, which shows that these pastoralists had great 
talent in composition. 

They also produced works of sculpture. A dozen stone figures, represent- 
ing animals or humans, have so far been discovered. None of them are 
very large (the tallest measures some ioi/£ in.); they were carved from hard 
rock, granite or limestone, and are polished smooth. Their proportions 
are lost in the mass, giving them a certain ungainliness without, however, 
diminishing their genuine artistic expressiveness. Among them are figures 
of cattle, with and without horns, rendered in a crouching position; with 
one gazelle and one ram the head only is reproduced. The human 
figures are of two types: some have owls' heads, and others have twin 
heads, recalling the two-faced Janus of ancient Rome. Most of the sculp- 
tures were found in the central Sahara (in the Hoggar and Tassili), the 
ram's head near Tuat. The significance of these stone figures is not known; 
they may possibly have been idols. Some of them were found by the 
Tuareg and put to use in their fertility rites. 

In the painted human figures, which are sometimes life-size, the profiles 
reveal different racial characteristics. Some of them are European and 
others negroid, but the majority have long straight hair and copper- 
coloured skin, and apparently represent an intermediate type similar 
to the modern Ethiopians. It seems probable that during the pastoralist 
epoch people of several races lived in the Sahara at the same time; some, 
the negroes, may have been slaves of the others. Slavery was a common 



COMPOSITE 
FIGURES 



Fie. 44 



Fig. 45 



RACES 



139 



institution among all primitive pas- 
toralists; they had a patriarchal soci- 
ety of a highly military character, and 
once their flocks reached a certain 
size, they were forced to keep slaves 
to look after them and perform other 
menial tasks. Judging by the paint- 
ings from the Tassili the pastoralists 
of the Sahara were already slave- 
owners. 

In their headgear, clothing and most 
typical physical characteristics the 
human figures of the pastoral period 
resemble the Fula (Peuhl), a tribe 
inhabiting the Sudanese steppe at 
the present day. They are also pas- 
toralists and are said to have intro- 
duced cattle-raising to West Africa. 
They are not negroes, for their skin 
is copper-coloured, their hair is long 
and not curled, and they are endow- 
ed with fine features; from an an- 
thropological point of view they be- 
long to the Ethiopian race. Various hy- 
potheses have been advanced about 
their origin and the route whereby 
they finally migrated to the Sudanese 
steppe. According to their own 
legends they came to Futa Jallon 
in the 8th century; from here they 
started out on their last migration 
east of the Senegal, which brought 
them to the Chad region. Previously 
they are said to have settled in 
Mauritania, where their ancestors 
were known by the name of Bafours. 
A people so numerous as our pas- 
toralists of the Sahara, who left such 
prolific evidence of their existence, 
cannot simply have disappeared 



140 



Herd of cattle. In the foreground one of the animals is being sacrificed. Pastoralist period. 
Jabbaren, Tassili. Length 10 ft. 




Masked male and female dancers. Only the men wear masks and dance ornaments on their arms 
and thighs. The same ornaments are worn by the 'White Lady' of Auanrhet, whidi belongs to the 
same school. The figure on the left is holding a wind instrument apparently made of horn. 
Period of 'round-headed men', with Egyptian influence. Sefar, Tassili. Length 2 ft. 10I/4 in. 



141 



institution among all primitive pas- 
toralists; they had a patriarchal soci- 
ety of a highly military character, and 
once their flocks reached a certain 
size, they were forced to keep slaves 
to look after them and perform other 
menial tasks. Judging by the paint- 
ings from the Tassili the pastoralists 
of the Sahara were already slave- 
owners. 

In their headgear, clothing and most 
typical physical characteristics the 
human figures of the pastoral period 
resemble the Fula (Peuhl), a tribe 
inhabiting the Sudanese steppe at 
the present day. They are also pas- 
toralists and are said to have intro- 
duced cattle-raising to West Africa. 
They are not negroes, for their skin 
is copper-coloured, their hair is long 
and not curled, and they are endow- 
ed with fine features; from an an- 
thropological point of view they be- 
long to the Ethiopian race. Various hy- 
potheses have been advanced about 
their origin and the route whereby 
they finally migrated to the Sudanese 
steppe. According to their own 
legends they came to Futa Jallon 
in the 8th century; from here they 
started out on their last migration 
east of the Senegal, which brought 
them to the Chad region. Previously 
they are said to have settled in 
Mauritania, where their ancestors 
were known by the name of Bafours. 
A people so numerous as our pas- 
toralists of the Sahara, who left such 
prolific evidence of their existence, 
cannot simply have disappeared 



140 



War chariots 
Fics. 41, 46 




without trace. Probably they were driven from the central Sahara by 
other invading tribes and forced to move further to the west, where the 
rock pictures they left behind them are indeed of more recent date. 
Moreover, they were compelled to look for better pasture-grounds owing 
to the desiccation of the Sahara, for which they and their animals were 
to a certain extent responsible. It is by no means a mere coincidence that 
the Fula (Peuhl) today inhabit the entire grass steppe of the Sudan, 
and it is very likely that they are the direct descendants of our Saharan 
pastoralists. The paintings and engravings which they left behind them 
are signposts along the route of their migration. 

The war chariots belong to two different styles: 'flying gallop' and 
schematized. The former is incontestably the older. It is found only in 
the central Sahara, the Tassili-n-Ajjer, and the Tassili of Ti-m-Missao. 
These chariots are thought to have been drawn from memory by people 
who had attended circuses in the Roman style in North Africa. A similar 
explanation has been given for the animal engravings of the bubalus 
and pastoralist periods: elephants, rhinoceroses, hippopotamuses and 
giraffes, which the artists may have seen when participating in caravans 
or campaigns in the Sudan. This thesis fails to take into account the large 
number of figures, as well as their distribution and antiquity, which in 
the case of the engravings is proved by the patination and in the case of 
the paintings by the resistance of the ochre colours. Surprising though it may 
seem, these chariots were used in the Sahara, just as the bubalus, elephant, 
rhinoceros and other animals were also once denizens of this area. 
The fact that the chariots 'at the flying gallop' are limited to the central 
Sahara shows the original location of the warrior peoples who used them. 
These were the tribes referred to by the ancients as Garamantes. The fact 
that several of these paintings were discovered near the spring of Ti-m- 
Missao, in the centre of the Tanezrouft, half-way between the Hoggar 
massif and the Adrar des Iforas, proves that the chariot peoples advanced 
far on their journeys of exploration and penetrated as far as the Adrar 
des Iforas, where an engraving of a chariot was discovered at the oasis 
of Arli. The southern latitude of these engravings suggests that these 
peoples even reached the Niger in the area of Gao. 

Any remaining doubts as to the use of chariots in the Sahara may be 
dispelled by the fact that they appear in hunting as well as in battle 



Fig. 44 — Idol with head of owl. Tabelbalet 



142 



Fig, j$ — Stone sculpture. Head of ram, 
viewed from ihe front. Tamentit, Tuat 



scenes and that there are also representations of ox-drawn chariots with 
two shafts which apparently served the purpose of carrying loads. In 
addition to this the authors of antiquity make frequent reference to 
the use of chariots among the Garamantes, and the chronicles of the 
16th Egyptian dynasty mention this in connection with the Libyans. 
Thus these chariots and the accompanying human figures in the double 
triangle style are evidence of the penetration of the white race into the 
Sahara and their expansion to the south along the mountain massifs. 
The schematized chariots, mainly depicted in engravings, are, as has 
already been mentioned, of an early date and belong to the equestrian 
period. These are much more widely diffused, for they are to be found 
not only in the central Sahara but also in the Fezzan, Mauritania, southern 
Morocco, southern Oran and north of the Tibesti. Soon after it first 
appeared in the central Sahara this vehicle must have conquered a vast 
area and been adopted by neighbouring tribes. In many pictures two 
or more shafts are depicted; these are apparently carts. 
Riding, which is the next stage after chariots, is only a modification in 
the use to which horses are put; and the horsemen are represented in the 
same style and carrying the same weapons. 

The area in which these equestrian figures are found is the same as 
that of chariots 'at the flying gallop', but the pictures are more numerous, 
indicating that this mountainous area was more densely populated. In 
this epoch the old Libyco-Berber written characters appear, frequently 
combined with equestrian scenes. The horseman equipped with javelin 
and round shield, a decorative plume upon his head, and finally with a 
knife dangling from his forearm, is not seen beyond the mountains of the 
Sahara but only as far as the Air and the Sudanese plain between the 
Adrar des Iforas and the Air. This corresponds roughly to the territory 
inhabited nowadays by the Tuareg. 

It was only much later that the horse appears to have been introduced 
into the west, that is to say into Mauritania, and to have been adopted 
by the population of that region; these people also differed from the others 
in regard to their weapons, since they carried spears and did not have 
a knife dangling from their arm. Furthermore, it seems that with the 
advance of the Arabs the use of horses became general, an assumption 




Expansion of the 
white race 



143 




The 'White Lady' or 'Horned Goddess'. This fresco, a splendid work of art, is painted beneath a 
small overhanging rock at a site that cannot have served as human habitation. The stippling on 
the body, shoulders, breasts and legs probably represents tattooing. The loin-cloth and dance 



144 






confirmed by the appearance of the saddle with a support for the back. 
In the east, .in the direction of the Tibesti and Ennedi regions, the horse 
remained unknown for a long time and is rarely represented; early Libyco- 
Berber inscriptions are also rare. The horse may not have reached this area 
from the north or the central Sahara, but from Darfur; here, too, its use 
became general only after the advent of the Arabs. 

We have already mentioned that the relatively recent introduction of 
the camel can be proved by the evidence of rock pictures. Engravings and 
paintings of camels are to be found throughout the Sahara. The simulta- 
neous appearance of written characters, the occasional association with 
the horse, and the differences in weapons and harness make it possible 
to classify these representations chronologically. 

Thus it can be concluded that the camel came from the north-east, for 
most of the oldest representations are to be found in the central Sahara, 
i.e. in the Tassili and Hoggar. From this area it conquered the south and 
west. More characteristic are the pictures in the Tibesti showing riders 
with a knife dangling from their arm, as is common among the Tibbu 
of the present day. The legends of these peoples also relate that the 
camel came from the east. The harness consisted of a primitive wooden 
saddle similar to that still used today by the Moors of the western Sahara; 
the side-saddle did not come into use until later, and its origin is 
unknown. The camel is nowadays found throughout the Sahara and it 
may justly be said that all life in the desert depends upon it — in so far 
as conditions have not already been revolutionized by motor transport. 
Thus we see that the engravings and paintings of the Sahara are veritable 
historical chronicles which enable one to reconstruct in broad outline 
the past of the peoples who lived in this area. 

The existence of rock pictures in the Franco-Cantabrian region and 
eastern Spain, as well as in South Africa, has given rise to the idea that 



Camel period 



AFFINITIES WITH 
ROCK ART OF 
EUROPE AND S. AFRICA 



ornaments, arm and knee bands, and the cuffs and tassels worn round the wrists and ankles 
consist of plaited fibres or thin leather straps. The horns on the head appear to be supporting 
a cornfield, from which seeds of corn are falling. This may be the goddess of an agrarian cult, a 
predecessor or copy of the goddess Isis, who was credited by the Egyptians with the invention of 
agriculture. The small figures painted in red ochre appear to be covered by the 'White Lady', but 
in reality they have been painted over this figure; the impression of transparent is due to the fact 
that the ochre would not adhere to the white pigment and faded. The 'White Lady' dates from the 
period of 'round-headed men', with Egyptian influence; the figures painted in red ochre date 
from the pastoralist period. Auanrhet, Tassili. Height 4 ft. 1 in. 



145 




Fig. 46 — Engravings of schematized 
chariots. Haut In Daladj, Hoggar 






oM) 



the engravings and paintings of the Sahara, occupying as they do a 
central position between these two regions, could be the connecting link 
in a large civilization of prehistoric artists spread over the two continents. 
Certain resemblances between the paintings of eastern Spain and those 
of the Bushmen have led some authors to speak of an affinity and to draw 
parallels which at first sight appear convincing. This theory was very 
tempting, but lacked any serious scientific foundation, especially from 
the chronological point of view. The decisive factor here is the age of 
the works concerned: is it really in order to attempt to establish an 
affinity in style between works separated from each other by as much 
as ten millennia or longer? Moreover, the rock pictures of the Sahara 
are so numerous that it is easy to find among them human or animal 
figures that bear a striking resemblance to those found in Spain or South 
Africa. The mistake has been to select examples quite arbitrarily and to 
neglect the broader issues which, had they been considered, would have 
changed the picture completely. 

It is true that these three art regions do have one thing in common — 
the naturalistic character of the works; but even if there are some affinities 
between particular styles, this does not prove that they can be traced back 
to a common origin. 

The antiquity of the eastern Spanish paintings is disputed, and the same 
applies to those of South Africa. Breuil used to hold the view that 
the art of eastern Spain, which is so very different from that of the 
Franco-Cantabrian area, belongs to the final phase of the Palaeolithic or 
at the latest to the Mesolithic. Spanish archaeologists nowadays plead for 
a later date, i.e. the Neolithic. According to Breuil the appearance of the 
bow is evidence of an African contribution to the eastern Spanish school, 
but in a purely technical, not in an artistic sense; and for a long time he 
believed that the influences which gave rise to the art of the Saharan 
pastoralists were transmitted in the reverse direction. But since the great 
discoveries made in the Tassili this world-renowned authority on pre- 
history has become more reserved in his opinion on this issue. 
When one examines the matter more closely, one finds that the oldest 



146 



engravings of the bubalus period in southern Oran have no demonstrable 
affinities with the engraved or painted rock pictures of Europe. Not only 
do these monumental figures only display very vague stylistic affinities Stylistic affinities 
with the art of the Franco-Cantabrian area, but it appears that religious 
conceptions played a part in the choice of these motifs of human figures 
in association with animals. How could it be otherwise when the two 
complexes of rock pictures are separated from each other by some 12 to 
15 millennia? Can one imagine an artistic tradition being handed down 
over such a long period of time? In any case such a transmission could not 
have taken place through the intermediacy of the eastern Spanish painters, 
for their style has absolutely nothing in common with that of the bubalus 
period. We are therefore of the opinion that the great bubalus engravings 
owe nothing to European art, and that the analogies to be found in the 
engravings are probably only accidental. 

The oldest paintings of the Sahara, those of the round-headed period, 
are purely negroid in style. It is beyond question that no affinity exists 
between them and European rock paintings, either in the Franco- 
Cantabrian area or in eastern Spain. The art of which they are the expres- 
sion is even more independent than that of the bubalus period engravings; 
it is inspired by specific religious conceptions and manifests such original- 
ity of form that it must be regarded as an indigenous creation and the 
source from which negro art sprang. 

The art of the pastoralists can at the most be compared with that of 
eastern Spain. In both groups human beings are frequently represented 
in action and the themes are the same: scenes of battle, hunting and daily 
life. In both cases it is profane art, to all appearances devoid of any Profane art 
religious conceptions. At first sight some relationship seems possible, but 
a more thorough examination shows that the art of the Saharan pas- 
toralists was far more highly-developed and considerably less conven- 
tional. What then are the affinities between this art and that of other 
regions and periods? 

Until recently the antiquity of the cattle period paintings in the Sahara 
was still uncertain, although they were believed to be products of the 
Neolithic epoch. But the latest discoveries in the Tassili have brought 
two reliable clues: the first of these is the age of two layers of deposit 
dating from the cattle period which have been subjected to radio-carbon 
analysis. The pieces of charcoal found in these layers, consisting mainly 
of kitchen refuse in the form of bones of cattle, date from the years 3550 
and 2450 B.C. As already mentioned, these are only average figures, 
from which it may be inferred that the first pastoralists came to the 

147 






Sahara about 4500 B.C. The second clue is to be found in the paintings 
of the Tassili, where river barques are shown very similar in construction 
to the barques of the Nile found in Egyptian pictures, particularly in the 
Nile barque rock drawings of the Upper Nile. They are classical Nile barques with 
the state insignia on the prow. Boats of this kind certainly did not 
originate in the Sahara, and the state insignia suggest Egyptian influence. 
If this is so, the pastoralists would have come from the Upper Nile area. 
This argument is reinforced by the human type represented and the 
remarks above concerning the origin of the Fula (Peuhl) tribe. 
The painters of the cattle period thus did not come from Spain, driving 
before them their bos africanus and bos brachyceros (which some scholars 
believe to be of Asiatic origin). We now have to consider the question 



Shepherd with a long rope, to which lariats are fastened on at equal intervals. At the end of the 

rope is a stake, which was driven into the ground. In the Sudan shepherds still today use this 

method of tethering calves. Pastoralist period. Sefar, Tassili. Height of figure 4^/4 in. 




whether the eastern Spanish paintings are really so old as they are alleged 
to be and whether they do not contain elements of African origin, as 
several Spanish archaeologists hold. This daring hypothesis must of 
course be treated with caution, and the whole problem will remain 
unsolved so long as the age of the eastern Spanish paintings is not 
reliably established. 

As further evidence of affinity between the eastern Spanish paintings and 
those of the Sahara, it has been pointed out that articles of clothing 
similar to the bell-shaped skirts worn by the 'Dancing Women' of Cogul 
also occur in the Tassili; however, in the Tassili this skirt does not 
appear during the cattle period, but belongs to the horse period, which 
began after the 12th century under the influence of Cretan art. This 
impulse from the eastern Mediterranean therefore rules out the possibil- 
ity that the paintings were introduced from Spain, and the gap in time 
is an additional argument against the thesis of a direct link between 
Spain and the Sahara. 

The question of the relationship to the South African paintings is equally 
far from solution. As in the case of the eastern Spanish paintings, their 
antiquity is still in dispute, especially as some of them appear to be older, 
while others are doubtless later in date; the latter are the work of Bushmen 
who still engaged in painting less than a century ago. Apart from the 
'White Lady' of Brandberg, an exceptional case, all the human figures 
in the South African paintings have negroid and Bushman features. From 
the stylistic standpoint they are unrelated to the works of the Saharan 
'round-headed men', but greatly resemble those of the pastoralists. As 
evidence of their religious and magic character Frobenius has pointed 
to the occurrence of plants with a religious significance, which is an 
important factor distinguishing the South African paintings from those 
of other centres of rock art, both in Europe and the Sahara. It has been 
held that the Saharan and South African paintings both originated from 
one and the same artistic centre, which thus gave rise to two schools. 
Such a centre has as yet to be discovered and one asks oneself whether it 
ever existed. 

In fact there is no denying that rock art is primarily a Eurafrican phenom- 
enon. Possibly similar causes may have produced similar results on groups 
of people who were particularly susceptible to a certain aestheticism of 
form and colour, for did they not also (apart from minor differences) 
use similar stone implements? This would also explain the existence of 
various zones of rock art, with the schools arising locally and developing 
independently of one another. 



149 



AFFINITY WITH 
EGYPTIAN ART 



Chronological 
differences 




A certain affinity has been established between the engravings and paint- 
ings of the Sahara and those of pre-dynastic Egyptian art — particularly 
in the case of the southern Oran engravings, where the figure of the 
ram with the disc bears a startling resemblance to the ram-god Amon. 
This was, as we know, a Libyan deity, venerated in particular at the oasis 
of Siouah; later this cult was adopted by the Egyptians. Were the rams 
of southern Oran ancestors of this deity, or was the reverse the case? 
Moreover, in the engravings of the bubalns period in the Wadi Djerat 
a large number of human figures with animal heads have been identified 
reminiscent of the animal cults of ancient Egypt, fn this case, too, the 
question arises which of the two centres influenced the other. But even 
though the antiquity of the bubalus period engravings in the Tassili 
cannot be established with absolute certainty, there is no doubt that 
these works do not date from historic times but, in the view of most 
experts, from a relatively early Neolithic. The oldest documents of pre- 
dynastic art, on the other hand, originate from a time which does not 
go back beyond the El Amrah culture. These are not rock pictures but 
figures painted on clay vessels or incised on schist or ivory plaques; they 
are all in a geometric style which bears no resemblance to our large-scale 
naturalistic engravings. Some authors, such as Petrie, consider that El 
Amrah art was influenced by Libyan elements. 

The semi-naturalistic style does not appear in Egypt before the El-Gerzeh 
period, and these figures resemble those of the cattle period engravings 
in the Sahara. The engravings discovered by the archaeologist Winkler 
in the Wadi Hammamat in Upper Egypt and ascribed by him to 'ancient 
inhabitants of the Nile' belong to this period. These early documents 
of Egyptian art are mediocre in quality and far removed from the perfec- 
tion attained in the cattle period engravings in southern Oran and the 
Sahara. The fauna depicted do not comprise any archaic species, and the 
bubalus is conspicuous by its absence. In Egypt there are no engravings 
with polished lines such as are to be found in the Wadi Djerat and in 
die In-Habeter (Fezzan); nor are there any figures of elephants, rhinoc- 



Fig. jj — Goddesses with birds' heads. Egyp- 
tian influence. Jabbaren. 9I/3 x I 4 1 /i '"■ 



150 




'Antinea'. A woman kneeling in ritual posture. The arms and breasts are painted. The head-dress 
suggests that this is an eminent personage; in front a motif can be identified reminiscent of the 
Egyptian pschent, a royal symbol. The human figure painted on the head-dress at a later stage 
dates from the pastoralist period, the small figure on the right from the period of 'round-headed 
men'. Post-pastoralist period, with Egyptian influence. Jabbaren, Tassili. Height 6 ft. 44 !n - 



151 



eroses, hippopotamuses, giraffes, ostriches, etc. Among the works that 
are most accomplished in point of technique, which must also be regarded 
as the oldest, is the striking picture of an elephant with ears drawn above 
the body like the outstretched wings of a butterfly. But this curious style 
is also known in the Sahara, particularly in the Hoggar, Fezzan and 
Tibesti. The dating is fairly clear, for it was repeatedly found superposed 
upon figures of domesticated cattle, and must therefore be ascribed to 
the pastoralist period. This style is thus of great value to us as evidence 
that the bubalus period art of the Sahara is older than that of pre-dynastic 
Egypt. 

And when we recall the rams with discs of southern Oran and the human 
figures with animal heads in the Tassili and Fezzan, we feel justified in 
concluding that it was the Saharan artist of the bubalus period who was 
responsible for giving birth to Egyptian art. 



i Aissala has been analysed by Barth. The engravings of this site were recorded by the 
author in 1934 and published in 1949. Cf. Investigaciones arqueologicas en El Sahara 
Central y cenlro meridional, Cuadernos de Historia Primitiva, No. 1—2, Madrid. 1949. 

2 The author has examined personally the Hoggar, Tassili-n-Ajjer, Air, Kaouar, French 
Sudan, southern Morocco and southern Oran. 

3 Les Pier res Ecrites, Paris, 1921. 

■4 Die Felsbilder Fezzans, Leipzig, 1937. 



»52 



THE ROCK ART OF SOUTH AFRICA 

BY 

ERIK HOLM 



When Europeans reached the southern tip of the African continent on 
their first voyages of discovery they found there peoples with three 
entirely distinct means of livelihood: nomadic pastoralists, primitive 
agriculturalists who engaged in a certain amount of cattle-raising, and 
hunters or food-gatherers pure and simple. These three modes of exist- 
ence may be regarded as phases in human evolution: it was as though 
in this part of the world the different means of livelihood had not yet 
superseded one another. 

The Hottentots and Bantu, who were pastoralists and agriculturalists, 
belonged to the Late Stone Age, those in the third group to the Early 
Stone Age. These hunters and food-gatherers had no metals or domes- 
ticated animals; they lived solely from what they could obtain by hunting, 
fishing or gathering natural products. To make weapons, implements, 
clothing and ornaments they had only the rock about them and the 
bones, sinews, hides and feathers of the game they killed. They also had 
vessels made from ostrich egg-shells in which to store the most vital 
commodity of all, water. 

These dwarfish people (the hunters and food-gatherers) had an infantile 
skull formation, graceful but powerful bodies, and remarkably small 
hands and feet; but they possessed keen senses and adapted themselves 
readily to their natural environment. They knew the habits of the 
animals and the effects of various herbs upon man and beast; and they 
could interpret the meaning of all the celestial phenomena for their 
most vital needs with far greater skill than might be expected of a mere 
'savage'. In particular they had a highly-developed instinctive sixth sense 
— a faculty absent in modern civilized man, whose instincts have become 
increasingly atrophied, but which it is not surprising to find in a people 
living in complete harmony with nature. This instinctive sixth sense often 
manifests itself in a manner which is little short of miraculous. Most 
striking of all is their ability to reproduce what their natural talents 
enable them to perceive. It is astonishing to find an old Bushman telling 
his European master that, for instance, he can sense on the side of his 



MEANS OF 
LIVELIHOOD 



Sixth sense 



•53 



body the approach of a herd of springboks on the far side of the hill, 
because the animals' dark stripes are transmitted to his own person. He 
may also explain that he can foretell the killing of game because he 
experiences a distinct sensation, as he walks along, of the blood of the 
animal he is going to carry home dripping on his heels. 
A people with such intensity of feeling and perception can scarcely be 
expected to possess powers of imaginative speculation. But this is the 
criterion which ethnologists are fond of applying when assessing the intel- 
ligence of a primitive race. They are surprised to find that these people 
are unable to count beyond ten, the number of their fingers. But this 
shortcoming is counterbalanced by a practical wisdom that is hardly to 
be met with among Europeans. Thus a minister once condescendingly 
asked a very old Bushman how old he was, thinking that he would hardly 
be able to count the score of his years, and was dumbfounded to receive 
the reply: "as old as my keenest disappointments and as young as my 
boldest dreams." 

It was only to be expected that Europeans who three centuries ago came 
across these curious people should have come into continual conflict with 
their way of life, their ideas and their incalculable pattern of behaviour. 
These 'savages' were in their eyes incomprehensible prattlers and idlers, 
even thieves and vagabonds, who could neither be employed in any kind 
of service nor be entrusted with the least responsibility. On hunting ex- 
peditions, however, they were of course indefatigable and indispensable. 
Artistic talents But the most astonishing fact about these prehistoric people was that 
they possessed a rich art of their own, a boundless passion for relating 
the most subtle tales, and a love of music, dancing, and ornamentation 
on their implements and their own persons. But all these manifestations 
of cultural awareness would have been inconceivable if they had not had 
distinct philosophy of life and an ancient system of religious beliefs. 
In Europe and in other parts of the world where the only evidence of this 
primitive food-gatherer culture is in the form of stone tools, it was 
assigned to the Upper Palaeolithic, i.e. to the last phase of the Old Stone 
Age. It is to these hunters that we are indebted for the rock paintings of 
Africa. 

At the southern tip of this ancient continent the prehistoric pattern of 
existence has been preserved unspoilt right up to the present day. Here 
it was possible for students of prehistory to talk to the men who created 
rock pictures, to get to know their way of life, watch their dances, learn 
their language and record their myths, legends and religious beliefs; these 
Myths could even be studied in relation to the expression given them in rock 

•54 




'Encircled' rhinoceros. Very advanced silhouette style, with 'stopping out' of inner surface. Frag- 
ment of diabase from a hill with pictures. Western Transvaal, site not known. Transvaal Museum, 
Pretoria. Half actual size. 



paintings. Thus a most complete and vivid idea was obtained of a vitally 
important stage of culture, of which all trace seemed to have vanished 
until, in our own day, signs of it were rediscovered in unbelievably impres- 
sive, mysterious — but mute — pictures in dark caves. This prehistoric art 
was, however, already known in South Africa for almost three centuries, 
although no one suspected its connection with European prehistory. 
In actual fact even today hardly anything is known about the deeper 
meaning of European cave art. It is simply taken for granted as perfect, 
grandiose and mysterious. Since we have no documentary evidence of 



155 



Prehistory in the 
living present 



The Stone Age 
artist is alive today 



The problem 
of dating 



the religious beliefs that motivated it, we approach it in the manner of 
someone who sets out to study Christian art without knowing anything 
about Christian doctrine. The interpretation given of cave art owes 
everything to the personal imagination of the observer. Surely it ought 
to be possible to attain a better understanding of the significance of this 
art by studying that of South Africa? 

Whereas prehistorians in Europe are necessarily concerned only with 
events that occurred many thousands of years ago, in South Africa we 
have the paradoxical situation that the study of prehistory is synonymous 
with the investigation of the living present. It is this that makes it so 
stimulating to deal with this rock art: one can hope to obtain results 
which not merely concern the present day but also the remotest past of 
mankind — in Europe as well as Africa. This presents a challenge to our 
whole outlook on history. 

Characteristically enough, although we know that this art was practised 
until recent times (and is even practised still today), we know very little 
about its date of origin. Some early travellers gave accounts of the men 
who painted the cave pictures and described their equipment. They are 
said to have carried with them, hanging from their belts, ten different 
kinds of paint in the same number of receptacles, shaped like antelope 
horns. 

A warden in one of the game reserves, whose duties include the protection 
of the last surviving Bushmen in the Kalahari desert, was fortunate 
enough to be able to persuade one of them to demonstrate the technique 
of engraving on stone. With the aid of a sharp stone tool, using his right 
and left hands alternately, the Bushman pecked out an episode involving 
a lion, an ostrich and a human being, adding on top of them, in the form 
of a circle, a number of antelopes. Whilst doing this he hummed and 
sang to himself, giving expression to his pleasure in what he was creating 
by repeated exclamations and peals of laughter. It is quite likely that 
he was following some legend with which we are unfortunately unfamiliar. 
It may be added in passing that the warden played a joke on some pre- 
historians by showing them this work and asking them their opinion as 
to its date. They gave evasive replies and finally cautioned him against 
"unlawfully removing a prehistoric work!" 

Most of the European scholars who have studied this art naturally em- 
phasize its Palaeolithic character, since they are used to dealing with 
Palaeolithic works at home and recognize the same traits in South African 
rock art. Scholars from South Africa, on the other hand, tend to ascribe 
the works to an extremely late date. No doubt subjective factors play 



156 



a certain part on both sides: the European is accustomed to the fact that 
from the Renaissance onwards Europe has exerfed a cultural influence 
upon the rest of the world, and subconsciously he assumes that the prim- 
itive artists 'colonized' Africa from Europe in a similar manner. The 
prevailing idea among South Africans, the descendants of recent 
immigrants, is that theirs is a young country which they themselves, so 
to speak, have discovered and brought under cultivation. They find it 
difficult to imagine that mere dwarfish creatures, whom they have come 
to know in a primitive state, should be able to look back upon such a 
long cultural tradition, and that they should even have had some part in 
creating the earliest paintings, including those in Europe. And yet this 
culture is very old indeed! 

In South Africa, as in Europe, prehistorians have discovered the most 
important finds along the banks of the rivers, which attracted the earliest 
inhabitants — mainly on account of the water, as one would expect, but 
also because of the pebble-stones, which they used to make tools. Here, 
embedded in heaps of broken stones on the terraces along the banks, 



Distribution of rock art in South Africa 



MA J OI'O* - M.4SSJF 




Elisabeth ENGRAVINGS O 

PAINTINGS • 



157 







Periodization 



Fig. 48 — Statuette of a young girl. Relief 
in calcareous spar. Sireuil, Dordogne 



together with the fossilized bones of the animals they killed, there have 
been preserved stone artefacts which were produced out of medium- 
sized pebble-stones over the course of many thousands of years. The 
different types of tool and the technological modifications that occurred 
in the utilization of the material — as well as the geological evidence, 
from the alterations in the flow of the river — make it possible to deter- 
mine the date of these prehistoric implements. 

Unless these old stone artefacts had been discovered, and the astonishing 
fact ascertained that they are sometimes identical with those found in 
Europe, no one would ever have thought that human settlement in this 
remote subcontinent dated so far back in time. But every year further 
evidence is accumulated to show that this region was settled by man at 
a very early stage — if not actually earlier than any other area. It soon 
emerged from careful comparisons of the oldest artefacts with those found 
in Europe, as well as in other parts of Africa, that — contrary to the 
original supposition that the impulse came from Europe — there took 
place here a homogeneous independent development, rooted in its own 
beginnings and not at all dependent upon European influences. The 
necessity therefore arose for a separate scheme of classification to be 
worked out for South Africa, governed by the alternation between arid 
and humid periods, as distinct from the alternation between mild periods 
and Ice Ages in Europe. 

But by and large there is a certain coincidence in the morphological devel- 
opment. For this reason, as is the case in Europe, the Stone Age is divided 
into three stages; but the periods do not coincide. In South Africa only 
the earliest phases are classed in the first stage, the 'Old Stone Age', which 
corresponds roughly to the hand-axe or so-called Chelles-Acheul culture 
in Europe. It is known as 'Stellenbosch'. 

The next phase is the 'Middle Stone Age', in which there appear new 
blade artefacts produced by a Levalloisian technique, in some way akin 
to the Mousterian in Europe, but showing greater refinement. Between 
the earliest phase and the Middle Stone Age are the Fauresmith facies 
and a stage of culture running parallel and known as the Sangoan. The 
'Late Stone Age' in South Africa would thus correspond roughly to the 
Upper Palaeolithic in Europe, i.e. the stage in which European art first 
appears. In South Africa this Late Stone Age in fact continues up to the 



158 



present day. The European Mesolithic and Neolithic can only be traced 
in minor details, such as, for example, the decreasing size of tools and a 
few polished artefacts. We are therefore primarily concerned with a 
Palaeolithic stage (according to the European classification). 
It is of particular importance to note that no interruptions took place 
in this evolution, that the thread was never broken, and that each new 
phase developed out of elements of the preceding one. This indicates an 
undisturbed organic growth, in particular extending up to the decisive 
culminating stage of the Middle Stone Age, the Magosian, a homogeneous 
culture diffused over the entire continent, in which the art of the south 
is born. Apart from this single, though important point, not nearly so 
much information for the history of art is derived from the study of 
tools as happened to be the case in Europe. Attempts to associate the 
development of art with that of artefacts, as in Europe, have led nowhere, 
and so far as South Africa is concerned it is necessary to follow a completely 
different line — one that will be set out here for the first time. 
What we have in mind may perhaps best be indicated by reference to 
the study of tools. From the start there appear certain types of tool with 
a specific shape: either a leaf-shaped 'pick' or a sort of 'spearhead', a beak- 
shaped 'scraper' or a 'disc'. But it is quite impossible to determine what 
practical purpose these types were designed to serve. They have therefore 
with reason been termed artefacts, a designation which throws light upon 
their most striking characteristic: they are fashioned almost like small 
works of sculpture. This technique represents an effort to give expression 
to an impulse to model plastically. It is thus difficult to avoid the im- 
pression that these oldest 'works of sculpture' were really attempts to 
fashion very simple idols, or alternatively abstractions of some kind. 
This need not rule out a practical purpose, such as a weapon; for in his 
undifferentiated state man does not distinguish between 'abstract' and 
'practical' spheres. Even in our own eyes a weapon means more than 
simply a 'lethal instrument'. For this reason we ought not to find it 
impossible to understand the Bushman who believes that an arrow-head 
of glittering crystal is made of the substance of the stars, and who, as he 
sharpens it, addresses to it the prayer: "may I hit my mark, for you never 
miss yours and are quite unerring!" 

This example should really suffice to show that not even artefacts can 
be fully explained by a purely technological approach, particularly at a 
time when a mania for technology seems to prevail over all spiritual 
matters. How then should such an approach be capable of disclosing 
the secrets of human customs and attitudes? It is equally inadequate to 



Uninterrupted 
evolution 



Art and artefacts 



»59 



'Magic' 



try to comprehend the spirit of a new-born form of art by reference merely 
to external needs, to culture materielle. But this is the domain into 
which the study of artefacts falls, and it is on this that the study of Stone 
Age art in Europe has been based. 

In Europe this occurred under strangely contradictory auspices: in the 
very caves in which excavations had been carried out for decades in an 
eager search for artefacts (as supposedly the sole evidence of prehistoric 
civilization), prehistoric art was discovered — by a child! And this mate- 
rialistic outlook was so deep-rooted that scholars were at first loath to 
admit that there was a spiritual aspect to Stone Age man, let alone that 
he was capable of creating works of art. 

When this great mural art had finally found recognition, and 'sketches' 
for the pictures were discovered as well as artefacts, it was doubtless to 
be expected that these artefacts should be regarded as the first stage in 
classifying this art. But here again undue significance was attached to the 
utilitarian aspect in approaching this question of artistic expression, 
which is after all a purely spiritual matter. The inevitable consequence 



Detail of a quagga in engraving and pecking technique, on diabase. Maretjiesfontein, western 
Transvaal. Various periods. Author's collection, Pretoria University. Slightly less than actual size. 



JE* ■*»'" 




160 



was that the concept of 'magic' was employed as an adequate means of 
explaining the 'purpose' of this prehistoric art. It was thought that in 
all 'primitive' cultures a specific belief had prevailed, in the light of which 
it was possible to explain the holes made by missiles and the 'arrows' 
added subsequently: the hunter was supposed to have painted figures of 
game on the cave wall to assist him in making a kill. In this way art was 
considered to have had the same 'purpose' as the manufacture of weapons: 
both merely served to provide man with food! 

In South Africa, fortunately enough, conditions were not so simple as all 
that, and it was therefore necessary to search for different principles in 
trying to understand rock art. 

Among the most important works are the engravings cut in the rocks Engravings 
found on exposed hill-tops situated over an extensive area in the interior 
of the country. At these sites tools could not remain preserved in strata 
undisturbed by time, as was the case in deep caves, but are found scat- 
tered about the surface, the earliest ones next to those of very recent 
origin. The archaeological picture thus resembles the ethnological one, 
as though in Africa successive periods had not superseded or overlaid one 
another — although the passage of time is clearly evident from the terraces 
formed by the rivers. Such formation of terraces was, of course, not possible 
on a rocky hill-top. 

A similar situation prevails with the other important pictures found in Rock galleries 
rock galleries in the mountain massifs where paintings have been 
preserved. Most of these sites are open caves (galleries) with sloping rock 
floors on which it was no more feasible than it was on hill-tops for cultures 
to be deposited in an undisturbed and stratified form. But even in the 
case of those few sites where deposits are to be found the question arises 
— and rightly so — whether it is at all permissible to regard these rock- 
shelters as habitations, or whether they should not rather be considered 
primarily as places of worship; and whether secular life, of which relics 
have been found, can tell us anything about the 'meaning of art'. 
When all the evidence is considered it must be assumed that these were Initiation 
places of secret initiation, access to which was prohibited to those not > — 
concerned in the rites. (It was for this reason that in Europe places of 
worship were located in the interior of dark caves!). It was only once 
people had begun to lose their awe of these sacred spots and the mysteries 
associated with them, and these localities were perhaps used merely as 
places of refuge, that the whole picture changed. But this change must 
have come about as a result of some extraneous influence, such as invasion 
by some new tribe who competed with the hunters for the means of subsist- 

161 



ence. In this case the bulk of the secular refuse can only date from a 
•/ period in which the sites had lost their original ritual significance. In 
South Africa there was no need to seek shelter in caves from the cold, 
and caves were not frequented until this region was invaded from the 
north by pastoralists, and later by primitive agriculturalists. Thus excava- 
tion at these sites would almost certainly give us an erroneous picture; 
the evidence of tools from the last phase would suggest that this art was 
of very recent origin. According to some scholars who have sought to apply 
undiscerningly European methods of excavation and investigation to paint- 
ings in the Drakensberg, they date only from the 17th century, when the 
indigenous peoples were driven back into these inaccessible mountain 
fastnesses by Bantu tribes migrating to the south. 

But can such a situation of utmost distress have been at all suited to 
artistic representation of an almost Elysian peaceful existence, as suggest- 
ed by the thousands of magnificent animal pictures preserved in these 
caves? Is it conceivable that the hunter peoples, themselves now hunted, 
persecuted and fearing for their lives, should have only developed their 
art at a moment of the gravest emergency? Only the last pictures orig- 
, inate from a period of such calamity; and they clearly reveal in the choice 
of subject the change that had taken place. For now these paintings depict 
battle scenes between the dwarfish hunters and their gigantic negro ad- 
versaries, who are shown carrying shields. Instead of peaceful hunting there 
is cattle-stealing, and the sacred places become refuges for rogues proudly 
vaunting their villainous exploits. These stirring little placard-like pic- 
tures merely illustrate the decline, the complete disintegration — as Breuil 
put it succintly, 'the total mobilization' — of an ancient hunter culture. 
They tell us as little about the real character of these peoples as do 
excavations. 
AREA OF Only one positive fact emerges from what has just been stated: in their 
utter despair these primitive hunters endeavoured to remain in their 
ancient area of habitation, sanctified by the existence of hundreds of 
sacred places. In the course of this struggle their whole character under- 
went a complete change: instead of men used to leisure and contempla- 
tion, endowed with a rich mythology, who cultivated hunting as a fine 
skill and regarded art as the supreme purpose of existence, they became 
vagabonds, rendered treacherous by their inferiority, who carried off as 
booty the invaders' cattle and waylaid them with poisoned arrows. 
Thus in these pictures we can follow the tragic fate of one of the most 
interesting peoples on earth — a tragedy all the more unfortunate because 
it was inevitable. It might be objected at this point that the diffusion of 

162 




Above, monochrome red cow antelopes; below, elands running towards the right; below, and on 
the left, small red human beings and animals. Sandstone gallery in the National Park, Drakens- 
berg Mts. 



this rock art — the paintings in inaccessible mountains and the engravings 
on the edge of the Kalahari desert — in itself clearly indicates that this 
race of artists was dying out. This common opinion has probably been 
suggested by the presence nowadays in this area of peoples with other 
modes of existence: pastoralists and primitive agriculturalists. But one 
has to bear in mind that the hunters actually preferred this kind of 
country because it was easier to find game there. 

In the case of the semi-desert country (such as the areas on the edge of 
the Kalahari have been from times immemorial) the twofold advantage 
is immediately obvious: the primitive hunters not only preyed upon wild 
animals, but were themselves sometimes the prey of these ferocious beasts. 



REGIONAL 
DISTRIBUTION 



Significance 
of the sites 



What chance would a dwarfish man stand against these creatures in an 
impenetrable thicket or in tropical jungle — quite apart from the exist- 
ence there of disease-carrying mosquitoes, against which he could not 
even protect himself by fire, his main safeguard against wild animals. 
An almost arid region with occasional water-holes was for him the ideal 
hunting-ground; at these few places the game was easily caught, for at 
some point thirst would drive the animals to the water-holes, where they 
could easily be ambushed. The hunter also has excellent opportunities 
along tracks through the mountain passes, particularly at the foot of steep 
precipices. A map showing the various rock picture sites, of which more 
than 2000 have so far been discovered, will serve to demonstrate how 
extensively diffused they were. It can be taken for granted that the hunters 
had the whole country to choose from when selecting their hunting- 
grounds. But they had very important reasons for confining themselves 
to certain specific regions. For their cult and the art connected with it 
they needed suitable sites, and these were to be found only where there 
were either protected walls for paintings or suitable rock surfaces for the 
execution of engravings (Map. p. 157). 
The following regions can be distinguished: 

1. the granite caves of the north, extending from the north of South- 
West Africa right across the subcontinent to Rhodesia and the northern 
Transvaal; 

2. the sandstone galleries, forming a semi-circle around the entire remain- 
ing southern part of the subcontinent; (here the paintings appear exclu- 
sively in their natural setting); 

3. the interior of the country, where the smooth vulcanic tufa, partic- 
ularly in the form of diabase, provided a tempting surface for engravings. 
This phenomenon is of special significance because it shows that to the 
primitive hunter artistic considerations seemed just as important as 
economic ones when choosing a suitable hunting-ground. It was one of 
the essential necessities of life for him to be able to express his ideas in 
the form of rites and art; without this he could not exist. Thus the 
impression of the primitive peoples gained by the Europeans on their 
arrival in the country, although based on their own observations, was an 
erroneous one; to obtain a proper picture, it would have been necessary 
to visualize them in their original state. 

In general rock art is envisaged even today as a more or less mechanical 
pastime, simply as a means of whiling away the time by decorating one's 
natural habitation. Although these Bushmen were sometimes met with 
in the open, leading an almost animal-like existence in primitive nests 



164 



Fig. 49 — Symbol of waxing moon: eland with stunted 
extremities, tyryburg, Bechuanaland (Transvaal Museum, 
Pretoria) 



made of grass, they had for the most part already 
fled into their mountain fastnesses. This led to the 
belief that the caves containing paintings were 
their regular habitations, which like ordinary 
tasteful civilized people they had decorated with 
nice little pictures depicting their daily envi- 
ronment and occupation. But originally, as has already been mentioned, 
the hunters used these caves neither as dwelling- nor as hiding-places, 
but for their devotions, since they credited them with mysterious magic 
powers; their remoteness and seclusion gave them a natural aura of 
sanctity. This is readily apparent if one thinks of the low hills bearing 
pictures in the interior of the country, which assuredly could not afford 
the least shelter against an enemy, and not even against inclement weather. 
The fact that these hill-tops, as well as cave walls, are literally covered 
with the most magnificent pictures, elevates them to the status of prim- 
itive natural sanctuaries; they are thus among the oldest sacred places 
of mankind. 

From the start Europeans studied the Bushmen and their art from the 
standpoint of social pathology; and this approach, like that of the ar- 
chaeologist, made it difficult for them to obtain a correct idea of the 
fundamental principles by which their art was governed. Thus doubt 
could be expressed as to whether such an inferior primitive people could 
have been responsible for the majestic ancient pictures which can be 
identified in strata and are older than the neglected ones of the last 
phase. Serious scholars who recognized that these pictures were of greater 
antiquity were reluctant to abandon the view that they could have been 
produced only by some extinct race. But hitherto it has not been possible 
to discover anywhere a find which could substantiate the claim that a 
connection existed between this art and forerunners of the Bushmen. 
The various racial types with Neanderthal characteristics must be ruled 
out, for the reason that even in Europe no artistic achievements 
can be ascribed to them. Boskop man, frequently put forward in this 




Racial type 



165 



connection, cannot be excluded, but with his much greater cranial capac- 
ity he displays unmistakable Bushmanoid characteristics. In this connec- 
tion the hypothesis was advanced that the dwarfishness of the present-day 
Bushmen represents a sort of infantile involution. In the caves on the 
south coast of Cape Province, where many interments took place, this 
burning question seemed to have been solved, since it was ascertained 
here that over the head and the upper part of the body of the deceased 
a stone slab was placed, the lower side of which sometimes bore paintings 
in the same colours as the murals. But these interred men had in the main 
San or Bushman features — although some scholars have established that 
the producers of a modified form of microlithic artefact (known as 'Wilton 
fades') had Cro-Magnon characteristics. These men were not buried in 




'Cloud' elephant, with rain pouring from its body. Silhouette style. Mount St. Paul massif, near 

Harrismith, Orange Free State. One-third actual size. 



166 



the manner described above, but had interred with them elaborately 
executed implements which pointed to the manufacture of paint. 
Excavations in Southern Rhodesia, South-West Africa and the Drakens- 
berg range yielded indications of a correlation between a particular type 
of tool and the paintings of the caves concerned; but the existence of such 
a link was accepted by some scholars and rejected by others according 
to their personal opinions, so that no final conclusions can be drawn from 
these finds. 

But in this connection an analysis of the psychological make-up and the 
traditions of the last pure Bushmen have yielded most astonishing revela- 
tions which are of fundamental importance for the whole of prehistoric 
art. Although these traditions have already been known for a century, 
strangely enough the results of these studies were not utilized as one 
would have expected. 

In the last century the British authorities in Cape Province were forced MYTHOLOGY 
to treat these uprooted primitive hunters as convicts. The German philol- 
ogist Wilhelm Bleek succeeded in obtaining from them by kindness and 
much painstaking effort a wealth of fables and tales which, taken in 
conjunction with their cults and rites, have made it possible to recon- 
struct a primeval mythology which throws light upon the earliest art of 
mankind. 

Each work gives us a glimpse of a world of ideas and beliefs of the deepest 
significance! 

Thus the Bushmen appear as the true upholders and heirs of a Weltan- 
schauung which inspired all the earliest rock art in Europe and Africa. In 
the light of this common spiritual foundation, relatively little importance 
need be attached to the many technical and stylistic diversities that exist 
within this single zone of primitive rock art. We shall now attempt, so 
far as is possible, to elucidate the position as far as South Africa is 
concerned. 

In attempting to establish the Weltanschauung of the Bushmen on the Weltanschauung 
basis of these profound and often obscure fables, one must begin by 
considering their metabolic approach: that basically there is no difference 
between matter and spirit. 

The fables, like the pictures, deal mainly with animals. But it is clearly 
apparent that these represent more exalted concepts. 
Just as the arrow-head can be a certain star, the powers of which are 
immanent in it, so also animals can symbolize other celestial bodies, as 
they have done since the earliest times in the constellations as we know 
them. At the same time these animals are also human beings, and each 

167 



fable begins with the preamble: such-and-such an animal was formerly 
a man, who belonged to the first tribe of men on earth. 
In the primitive state there was apparently no differentiation, and no 
special significance was attached to the individual living being or material 
object: a man can be a stone, just as a stone can become a picture, a 
spirit; everything can take place in heaven as well as on earth. 
The finest picture of this unity of all existence is given in the fable of 
the origin of the Milky Way: 

A girl threw glowing ashes from the evening camp-fire — which was an 
offence — up into heaven to enable the distant hunters to find their way 
home in the darkness. These ashes form, as it were, a link between the 
terrestrial and celestial path. 

The celestial luminaries wander about on earth unrecognized in human 
and animal guise, and each slain animal may contain some secret of the 
cosmos within itself. 
Mantis In particular, the mysterious 'praying mantis', which even the Greeks still 
associated with mantis, or soothsayer, is associated with an exalted divine 
miracle-working force. He is even believed to be the creator and guardian 
of life; he has his favourite animals, especially the eland, cow antelope 
and gemsbok, which he protects, in which he becomes incarnate, and 
which can then embody the stars, the sun or the moon. 
For the cardinal belief is that a living creature can be transformed into 
a higher being: even Mantis, the Creator, submits to death in order to 
reveal himself in fresh glory as a celestial body. The sun, the moon and 
every living thing is naturally subject to this eternal cycle of death and 
re-birth. In one of the principal fables, for example, Mantis is transformed 
into a cow antelope, who feigns death because it wants man to dismember 
it and deliver it from life. But soon the severed limbs slip from the hands 
of the terror-stricken men, and with the first rays of the sun are joined 
together once again. And now this being soars up to heaven in its true 
essence and circles the earth with a dancing step. It is now no longer 
animal, insect or man, although still recognizable as such, but a being 
comprehending the Universe, representing totality and the source of life, 
the very day-star. In South African art this Protean being is portrayed 
with exactitude in its various manifestations; but it is also to be found 
in the so-called 'magician' in the French cave of Les Trois Freres, and 
this in turn is the prototype of many later cosmic gods and demons, 
which speed across the sky like the day-star, with a dancing step. 
Dualism Another important factor is the dualistic principle immanent in the whole 
of creation. The Bushman considers everything that is plump, weak and 



168 



S3B 




Two polychrome giraffes, surrounded by human figures carrying loads and a 'tent', over an area 
densely covered with small painted human and animal figures in several layers. A little higher, 
taller men with javelins; on the right, a recumbent figure. Above, a white demon wearing a mask. 
Detail from the granite cave of Silozwane, in the Matopos. Large figures almost actual size. 



small to be female. Thus the full moon can only be female — but it also 
has an infantile neuter phase and even a male one. But in accordance with 
the idea of metabolism the moon can also be a girl or an animal, in 
which case rudimentary legs indicate growth. When full the moon 
represents a mature woman, or, more especially, that magnificent great 
animal, the eland. 

The new moon can be symbolized by a feather thrown into the air, the 
wish-bone of a blue crane, or the horns of various animals. 
From these, or from the blood-stained feathers of a slain ostrich, a shoe 
thrown into the air, a piece of hide cut into ever smaller pieces, the 
praying mantis causes the moon to rise again, slowly and well-guarded, 
from the waters of a lake. Each time this is preceded by death; yet the sun 



169 



restores everything to new life; so also does the moon, in its various 
Fig. 48 guises, behind which shines forth the image of an attractive and well- 
groomed woman. 

In the narrow cave of Sireuil in the Dordogne is a small picture of a little 
girl with rudimentary arms and legs stretching herself and 'blossoming 
out'. This and other early French pictures have their exact counterpart 
in South Africa — as, for example, in the eland from Vryburg in the 
Transvaal Museum. Here, too, the animal's legs are still rudimentary, 
corresponding to the waxing moon, although the head is not only fully- 
Fig. 49 grown but shines in the bright rays of the sun. 

What might otherwise be regarded as a badly-executed or unfinished 
figure, or even as one worked in two different techniques, is in fact clearly 
a well-thought-out illustration of a profound and beautiful mythological 
concept. And it is only from this mythological aspect that we can un- 
derstand the significance of the various motifs, and how it is that, although 
seemingly remote from each other, they yet have the same meaning. 
Many more examples could be adduced from South and North Africa, 




Fig. =,0 — Large painting representing a rain myth. From Philipp Cave, near Ameib, 
Erongo Mts., South-West Africa. After H. Breuil 



170 



is well as the Franco-Cantabrian area, to show how underlying myth- 
alogical concepts, as well as Bushman fables, have been illustrated in 
rock art. Everywhere this art is based entirely upon mythology. 
This uniformity was not thwarted by the fact that the fauna in the north 
differed from that in the south, or that climatic and material conditions 
varied: for example, in Europe there were no antelopes in primeval 
mythology, but their place was taken by other appropriate animals, 
rhere is a beautiful fable (rather similar to our tale of Little Red Riding 
Hood!) which Bleek's daughter Dorothea has made known from her 
Father's notes: 

VTantis (the praying mantis), which must be imagined as male, fatherly, Legend of the rain 
;ven avuncular, like a wise old hunter, fetches for his little sister, the 
foung springbok, some wild bees' honey from a hole in the ground which 
las been dug for this purpose. After being asked repeatedly whether it 
itill liked the honey, the little springbok suddenly stopped answering. 
Instead a fearful trumpeting resounded from above. When Mantis 
:rawled out of the hole to see what had happened, he discovered that a 
rerd of elephants had approached stealthily, abducted the child and 
exchanged for it a baby elephant. Enraged, Mantis kills the ungainly 
Daby and goes to look for the little springbok. Eventually he espies its 
•ed body as it frisks about nimbly and merrily in a broad valley, among 
:he slow-moving elephants, and he rejoices at the sight. But when he 
ipproaches, the elephant swallows up the little springbok, and when 
Mantis tries to climb into its mouth threatens to spit at him. Then Mantis 
juickly slips into its navel, wraps up the child, and lacerates the elephant's 
ntestines. The other elephants come to its aid with sticks and spears; 
aut instead of coming out at the navel as they expect, Mantis emerges 
:rom the elephant's trunk, carrying the baby springbok, and stabs the 
nonster to death. 

It is not too far-fetched to recognize in this tale the mythological descrip- 
ion of a thunderstorm. The nimble springbok with its red stripes 
epresents sheet-lightning, Mantis lightning, and the gigantic elephant 
:he cloud, rent asunder by Mantis. 

Pictures that clearly illustrate and substantiate this interpretation are to 
De found in all the regions of Stone Age art. 

n the plate on p. 166 there is a silhouette figure from Mount St. Paul 
n the Drakensberg range, showing an enormous misshapen elephant, 
which may be identified as a cloud, emitting streams of rain. The ren- 
dering in pure silhouette form is simply explained: the gigantic and 
shapeless form of a cloud is here illustrated by the massive towering 

171 



figure of the elephant. From its extremities, which are sketchily indicated 
— and even from its unfinished trunk — there are fine brush-strokes 
suggesting rain. 

From the opposite end of the country, from Philipp Cave, near Ameib in 
the northern part of South-West Africa, comes Fig. 50, depicting an 
elephant in white with a red springbok 'in its belly'. Breuil has rightly 
pointed to the close stylistic affinities with the oldest European figures, the 
elephants of Baume-Latrone (Gard department, France), where the 
curious distorted perspective of the head corresponds absolutely to that 
at Philipp Cave. The same nebulous manner of representation clearly 
suggests the cloud-like quality of the mythological elephant. It is not 
surprising, therefore, that in all the European mammoth figures, with 
their inordinately long 'hair', which sometimes shoots out arbitrarily 

Eland, in three colours, with small figures venerating it. Roof painting in the gallery near Heren- 
veen, east of Chrissie-Maars, Drakensberg Mts. Width's ft. 3% *Ȥ 






-m&. 



**-* 








&&** 



\. 



% "5* 



from the body in straight lines, and generally even covers their legs 
completely, conjures up the idea of a great rain-cloud moving along 
silently. 

There is even what might almost be called a literal illustration of the 
motif in the tale mentioned above engraved on a bone from the floor 
of the cave of La Madeleine in the Dordogne (Fig. 51). One can see 
quite clearly the insect entering into the gigantic animal's navel, the 
animal stumbling and being strangely transformed, curious objects 
pouring forth from its ear, 'horns' growing out of its head, its trunks 
disintegrating, lightning striking before its eyes, and (NB!) rain even 
beginning to pour forth from its tusks. It surely cannot be thought that 
these vertical lines simply indicate hair. 

'In the belly' one can identify a bison calf, a substitute for the little 
springbok. The foreshortening of the latter figure is typical of the symbol- 
ic manner of representation; this is found very frequently in rock pictures. 
This fact alone shows that abstract ideas are a basic element in the 
mythological outlook of primitive man. 

In North Africa, too, where the art is apparently so different, the same 
mythical subject-matter has frequently survived: in the Djebel Bes Seba 
(Algeria) there are elephants festooned with clouds (Fig. 53), and between 
them is the slender spirit-like springbok. In Ain Gudeja, near Asia, the 
cow elephant can even be seen 'swallowing up' this small animal (Fig. 53). 
Thus, as this example (which could be supplemented by many others) 
shows, the mythical element is to be found wherever rock art occurs. What 
has been said here makes it necessary for us to revise our conceptions 
about the 'purpose' of rock art: it will no longer suffice to assume that 
the portrayal of an animal figure was simply motivated by utilitarian 
ideas, even 'magic' ones. The animals are not thought of as game; in 
South Africa, for example, the springbok — the most popular quarry, 
among Bushmen even today — is very rarely depicted, and if so then 
always bearing a distinct relationship to the myth mentioned: it is rep- 
resented as a young animal nimbly leaping or playing about. It ought 
perhaps to be emphasized once again that all animals are characterized 
from a mythological standpoint. 

Nor does there seem to be any basis for the view that the animal figures 
are 'totems', i.e. that a specific animal may be the idol of a particular tribe 'Totem' 
or clan. The eland, the favourite animal of the sun (Mantis) is to be found 
throughout the whole vast realm of South African rock art, in all phases, 
and in drawings as well as paintings (in the latter case it is almost the 
only subject treated in the mountain galleries). This can only point to 



173 




Below, old remains of animals in outline; blurred monochrome silhouettes; on the left, head of a 
kudu cow in good condition. In the centre, line-drawing of a zebra. Above these, smaller figures 
in red and a demon holding javelins or switches. Right-hand wall of granite cave of Nswatugi, 

Matopos, Southern Rhodesia. Length of zebra approx. 131/4 in. 



Cosmic symbols 



the fact that the eland had a hierarchical significance in a system of beliefs 
common to all primitive hunters. From their mythology it is clear that it 
was an embodiment of the moon, which plays a vital role in their life. 
Of all natural phenomena it was the moon that was venerated most, and 
this magnificent great animal, the favourite creature of the giver and 
taker of life, occupied a special place in the heart of the primitive hunter. 
In other regions and ages the ox took the place of the eland as the symbol 
of the moon. 

The primeval form of this cosmic symbol is to be found in the myth of 
the 'killing' and 'resurrecting' of the evening-star by the sun. This is 
connected with the sacrificial slaughter of animals, which can also be 



•74 



interpreted in the sense that only the creator of all things, the day-star 

itself, can take away again this animal's life. 

Leo Frobenius studied carefully among the Pygmies of Central Africa the 

rites connected with this all-embracing belief in the everlastingness of 

the cosmos. These correspond with their myths, for at the first ray of dawn 

the Pygmies drew a picture of the animal they were going to kill, as though 

at the moment of sunrise this act created the animal and through its 

image placed it under the sun's spell; on the morrow, after it had been 

killed, it would then rise again, in a higher sense — after performance of 

the rite of covering the picture with the dead animal's blood and hair. 

It is thus not, as has been thought, a question of securing a substitute 

when no game was caught, but of the animal returning to the 'Father' 

who is the preserver of all life. The animal figures therefore represent 

immortalizations or deifications and, as it were, enter into the heavens as 

eternal stars. 

Another point that is often overlooked is that once the rites have been Rites 

performed and sacrifice made, there is no further desire on the part of 

these primitive hunters to preserve the animal figure. If it is not destroyed, 

they fear that this might adversely affect their fertility. It is for this 

reason that the pictures are placed in subterranean refuges, caves and 

other secluded places. Or at least signs of destruction have to be made on 

the picture, so that the animal should no longer 'walk about'. Thus we 

have here yet another clear indication of belief in transcendentalism, in an 

after-life of the soul. 

We will now turn to consider some examples. EXAMPLES 

In the Matopos, a massif of granite cones south of Bulawayo in Southern 

Rhodesia, we find magnificent caves as large as cathedrals, which form 

deep semi-circular recesses in the glittering blue rock. The beautiful view 

which they afford over a splendid landscape of acacias suggests that this. 

must have been a veritable paradise for hunters. On the walls in the 

interior of these caves we find this paradise most artistically represented 

in exquisite pictures, with a profusion of wild animals. The modern 

observer, who can no longer appreciate the mythology here represented, 

is content with this superficial visual experience. He hardly expects 

more from these animal figures, with their unusual colouring and 

consummate draughtsmanship, portrayed in animated movement. The 

cool of the vaulted cave, the trickling of a rivulet nearby, the softly 

shimmering twilight: what could have a greater emotional appeal? And 

yet all this conveys nothing of the motives that brought the primitive 

hunter to these caves. 



•75 




F*£- S 1 ~ Rain myth, as in Fig. 50, on ivory. La Madeleine, Dordogne 



For him the great mystery lay in the rock itself: it was the rock, he believ- 
ed, that could transform this 'chamber of the dead' into a vaulted heaven. 
To the walls of this glittering blue chamber he entrusted his ideas of the 
cosmos: with bold strokes he embellished them with his conception of 
the universal and the eternal, in the same way as the girl cast the spark 
from earth up to the stars in heaven. 

Nswatugi The plate on p. 1 85 depicts a fairly large section from the wall of the cave 
at Nswatugi. One can almost sense the vaulting of the cave in the curved 
bodies of the animals, which seem to be running along the 'horizon' — 
rather small reddish-brown silhouettes above an indecipherable jumble 
of countless superimposed paintings in dark and lighter layers, similar 
to those found in the engravings below the 'magician' at Les Trois Freres 
(France). There are so many paintings on this lower part of the light 
granite wall at Nswatugi that it is completely obscured by them; it is 
in vain that one seeks to make out the bodies of the animals, which 
appear to merge into one another. Between all these figures, each of which 
is a perfect drawing in its own right, is a mass of animated human sil- 
houettes, red and dark-coloured, and accurately portrayed — a teeming 
mass of bodies. 

ate p. 187 But upwards into the blue vault there stretch two great giraffe figures. 
Depicted in exactly the same solemn attitude, these fine slender animals 
stride along, dominating the whole vault (cf. plate on front cover of 
slip-case). Below them, at an angle, is a third giraffe, which also stands 
out from the mass of other animals; the galloping motif, accentuated 
more than in real life, conveys a greater impression of movement. 



176 



At first sight one can identify three different versions of one and the Three giraffes 
same motif. 'The giraffe in the centre (depicted on the front of the slip- 
case) is portrayed in a more rigid manner, with less attention to detail, 
almost like a simple archaic silhouette. The dorsal line is overlaid by 
another, somewhat darker colour, thus accentuating the head and legs. 
The spots are less conspicuous than on the giraffe painted a little higher 
up the wall; striding along in step, this animal is smaller, but represented 
plastically with classic poise. In both pictures the modelling and colour 
effect is produced by the use of only two colours, the lighter areas being 
left untouched when the second coat of paint is applied. 
The third picture, the most recent, has more than classical repose and 
stateliness. The animal is depicted galloping along, with its tail swinging; 
its legs are raised slightly higher than in the case of the other two animals. 
The general effect is enhanced by a new vision of colour, which makes 
the body appear vivid and bright, almost luminous. This picture shows 
greater maturity; the rigid, composed and static character of the other 
pictures is here relieved and relaxed. This is particularly apparent when 
one compares it with the monochrome silhouettes of small animals upon 
which the giraffe picture, with its white ground, is superposed. A sense 
of distinction and elegance is evoked by modelling with finely-ground 
pigment. It is not the neat bichrome spots on the body that set the tone, 
but the delicately shaded modelling. When compared with other styles, 
it appears as though these three specimens on one wall represent three 
definite phases of development; and indeed, this gradual evolution 
suggests that each period may have lasted for a considerable length of 
time. But it is impossible to estimate for how long, undisturbed in this 
secluded location, the infinite mass of indecipherable pictures below 
the three main figures continued to accumulate. Perhaps they bear some 
relation to the three giraffes? But this question is not what mainly interests 
us at the moment. 

What we are chiefly concerned with is the striking phenomenon that, 
despite their different dates of origin, the three versions of the giraffe 
motif correspond in such a striking way. This parallelism must have some 
significance. It would be easy to attribute it to emulation, and to say 
that the older picture must have served as the model. But we have to 
consider that the later versions are superior to the prototype and therefore 
could also have brought innovations in the way of motifs. We must thus 
reckon rather with a sanctioned version of an 'essential' motif, as is to 
be found later with pictorial representations in temples and churches, 
where the model is laid down and defined by specific religious conceptions. 




Rhinoceros, on western side of a rock; on the eastern side there is a lion. Below, an older small 
rhinoceros with dark patina. Summit of hill at Leeufontein, western Transvaal. Length of main 

animal approx. 2 ft. 6 in. 

Constellation What the basic idea in this case was may perhaps be discovered from 
information furnished by the Tati Bushmen, who identify the constella- 
tion of the Southern Cross as a giraffe. Taking this constellation as the 
body, and the surrounding stars as the legs and head, the figure of a 
giraffe striding along can be reconstructed — just as has been done since 
time immemorial with many constellations in the northern sky, which 
have been interpreted as animals. The image of the giraffe was evolved 
by primitive man from the idea of the cosmos evoked in his mind by 
the cave, and from the transcendental meaning of the religious mysteries 
which were revealed to him there. 

One technical archaeological point should not be overlooked in this 
connection, since it has a bearing on the age of these pictures, i.e. the 



178 



phase of development from which they originated. Both in the cave at 
Nswatugi and at Bambata nearby cultural deposits were found which 
positively invited excavation. At Bambata, in a mine of cultural informa- 
tion almost 20 ft. deep, there were found, in addition to weapons, blade 
artefacts, mostly older and dating from the African Middle Stone Age, 
which pointed to great antiquity. The tendency was to regard these 
cultural deposits, which were rightly taken as belonging to an early 
period, as coinciding in time with the predominantly monochrome paint- 
ings in the cave at Bambata. However, it has also been held that pictures 
from this very early period could scarcely have been preserved owing to 
the constant flaking of the rock wall. 

But on closer examination it clearly emerges that on these principal Ritual destruction 
animal figures some parts of the paintings have been destroyed. This 
can be explained with relative ease: the figures must have been damaged 
and defaced deliberately, for the missiles or blows which caused the 
damage were aimed at the animals' heads! In the case of all three animals 
it is only the head that has been damaged. On the other hand, it is 
precisely where the superposed paintings form a compact layer that no 
flaking took place. If weathering left its mark anywhere at all, then it 
did so on these layers bearing many superposed paintings, for these are 
doubtless of the greatest antiquity. 

In the case of our 'constellations' we are therefore faced with ritual 
destruction, an act which should not be considered divorced from its 
cosmic significance. If we recall to mind the blurring of pictures in the 
cult of primitive hunters (or the myth of Mantis, who, in the guise of a 
cow antelope, desires to be put to death), we see that this is a reflection 
of 'cosmic death and rebirth', the idea of the regeneration of the universe. 
It is in any case not a matter of killing the animal in effigy with the aim Regeneration of 
of procuring game, but a sacral safeguarding of one's own vitality against <he unwerse 
influences from the beyond; and this very fact indicates that these animal 
figures symbolized supernatural forces. 

It is therefore not surprising to find that damage was not done to every 
picture — for instance, most of the smaller pictures at Nswatugi were 
spared. We are sufficiently familiar with the psychology of primitive 
hunters to realize that we are not only dealing with pictures of transcend- 
ental import. 

With his mundane eye, so to speak, the hunter accurately observes the 
animals' proportions, and so is able to capture their likenesses with 
brush and crayon with the same skill that he shows in shooting them with 
his arrows. He strives to obtain verisimilitude, to come closer to their true 

179 



essence — as he conceives it; and this is far from being equated with mere 
terrestrial existence. His 'naturalism' is wholly remote from this world: 
in the most animated figure in the group of giraffes, the one on the right 
at the bottom, on the horizon, no attempt has been made even to show 
its spots; this served to emphasize all the more its vitality, its essence. 
We shall come back to this picture again in due course. 
The giraffe If we now consider the giraffe figures in other media of South African 
in engravings voc fc artj j n tne engravings in the interior of the country, we again find 
very different stages of development. In point of style our 'classical' 
bichrome version corresponds completely with the fine engraving in the 
Transvaal Museum at Pretoria. Both the colours of the tufa, the purple- 
brown colour of the thin oxidized coat and the light blue of the core, were 
utilized to obtain in two layers something approaching the same bichrome 
effect as is found in the 'classical' version. 

Also with a completely different engraving technique, pecking out with 
a pointed stone tool, the contours are neatly incised and the spots in the 
animal's coat 'stopped out'. The head is conceived as a single unit of 
colour, and the hind legs, which are 'in the shadow', are executed in a 
darker colour. 'Stopping out' the surface also serves to convey the visual 
impression of overlapping, and thus to produce a certain modelling effect 
in the area of the eye. As in the 'classical' version, this long-lined animal 
with its great neck is rendered in the same superb attitude of static repose. 
Morphological The motif must, however, be explained by a completely different mor- 
expenence phological experience: the stone stands erect like a dolmen, and the 
animal figure is accommodated to it. This stone offers such an inducement 
to plastic modelling that the animal's mouth even extends round the 
edge of the stone. Instead of being transferred on to a large surface curving 
inwards, like the sky, the figure bends outwards around the curvature of 
the stone and strains upwards with its long neck as though cast up into 
the air. But the underlying genetic basis is still the same: in one case it 
is the vaulting, in the other the shape of the stone which seems to form 
a link between earth and heaven, in the same way as the glowing ashes 
which the girl threw up into the sky to form the Milky Way. 
In so far as this idea was suggested by the location of the stone, rising 
up on the summit of a small hill, the artist's conception reaches out beyond 
the limits of this world. Both versions, despite their differences, therefore 
have the same purpose. This is the decisive point: it is no mundane vision 
that gives this artist his inspiration. One might suppose that the upraised 
head indicates browsing on a tree-top, but there is no evidence in rock 
art for such an interpretation. The motif is profounder, and must be 




Giraffe on upright diabase cone. The outline and the spots on the animal's skin are pecked out. 
Western Transvaal. Transvaal Museum, Pretoria. Height of animal approx. 2 ft. 3 in. 



interpreted in terms of the transcendental emotions that underlie the 
whole of this art. 

That it is indeed an essential feature of rock art that the living picture 
should step forth from the stone, so to speak, may be illustrated by 
another engraving, from the top of a hill bearing pictures near Leeufon- 
Plate p. 178 tein in western Transvaal. Here the rough-edged surface suggests the 
Suggestion armoured hide of a rhinoceros, this resemblance being still further 
enhanced by the sharp ridge of the rock. In this instance, too, the body 
of this great pachyderm is only pecked out in outline, with the ribs 
barely being indicated, in order to accentuate the two-dimensional effect. 
As in the case of the giraffe, the head is executed in detail, and thus given 
special prominence. This figure is also depicted with deeply incised horns 
(but on the other hand with a raised tail) and extends around the edges 
of the stone. 

On one of the lateral planes there appears the figure of a lion, in the 
shadow and scarcely visible; and below the stomach of the rhinoceros is 
a very archaic small animal, with a dark patination, which can be 
identified as another rhinoceros. On the large rhinoceros the outline 
seems to have been retouched in some parts at a later date, which may 
indicate the significance of this hill-top picture. 

As is shown by the points that correspond, the rhinoceros picture also 
originates from a classical period of art; it is vigorously rendered on a 
large scale and represents a vital mythological concept. The rhinoceros 
must no doubt have held a high position in the animal hierarchy; on 
every hill that bears pictures it is rendered in a prominent place, usually 
on the largest and tallest stone, next to the figure of the lion. And here 
we come to a new rule of this art, which shows that the animal figures 
stood in the closest relationship to the emphasis laid in the rites on the 
Cosmic significance cosmos: the figure of the lion is the first to be struck by the rays of the 
morning sun, thus showing (as Frobenius already discovered in North 
Africa) that the radiant lion is related to the day-star. The fearsome dark 
rhinoceros, on the other hand, must personify night, when the monster 
is especially unpredictable and dangerous: its image on the hill-top faces 
west, i.e. towards the setting sun. 

The full effect of this picture can only be appreciated when the moon 
is rising, and its horns shine like the 'horns' of the new moon over the 
horizon. It is in this effective manner that these pictures give expession 
to the most profound mythological concepts! The demon of night and 
darkness — from which the 'horns' of the waxing moon stand out, bringing 
consolation to the onlooker — is an impressive mythological symbol. (To 



it a crude humorous touch has sometimes, as here, been added: the tail 
is raised for excretion!). The dark demonic side in the imagination of 
primitive rnan was also represented pictorially in the figure of the angry 
and violent rhinoceros. These rhinoceros pictures are often invested with 
a fearsome majesty, as, for instance, when the animal is depicted in the 
act of frenzied attack. 

The interpretation given here of the sinister character of the rhinoceros 
is confirmed in a remarkable manner by the well-known figure in one 
of the lower mouths of the cave of Lascaux in the south of France, which 
has been the object of a good deal of misleading comment. This painting, 
situated in a secluded spot, represents an analogy of a Bushman fable that 
has several versions and treats the theme of total darkness: contrary to 
the sacred order of the cosmos, the moon (in this case a bison instead 
of an eland) has been killed, not by the sun, which alone would be her 
right, but by the rhinoceros — and, moreover, before daybreak. Out of 
revenge Mantis pierces the spleen of the dead animal, bringing darkness 
upon the world. It falls both upon the murderer and upon the God of 
Light, but the latter causes the new moon to spring like a bird from a 
feather thrown up into the air. 

In the picture at Lascaux the rhinoceros, the cause of this evil, is sprinkled 
with black drops of bile at it flees away. This theme is frequently met 
with in South African engravings, thus furnishing further evidence of the 
fact that the whole of Eurafrican rock art has a common mythological 
foundation. 



Fall of darkness 




Fig. 52 — Rain myth, as in Fig. $0, from Djebel Bes Seba, Saharan Atlas. After 
H. Obermaier and L. Frobenius 



183 



Mythological 
explanation 



It is only after one has elucidated these mysterious affinities, with the aid 
of living oral tradition, that one can appreciate the full spiritual subtlety 
of such a beautiful gem as the picture of the rhinoceros on a purple 
Plate p. 155 fragment of diabase from western Transvaal (now in the Transvaal 
Museum). This animal seems to be hampered in its movements; apparently 
blinded, it stamps about and sniffs the ground. This has been interpreted 
as though it were drinking at a water-hole, but the crudely chipped circle 
which seems to hold it captive means something more than a mere pool 
of water. The natural vesicles in the rock above the animal's head 
suggest, as in the examples just mentioned, the source of darkness from 
which the bile spurts forth, enshrouding the universe in darkness. As 
with the picture at Lascaux, drops of black fluid can clearly be seen 
spurting forth from one of these vesicles, forming a circle around the 
animal. 

In contrast to the technique hitherto applied, the entire contour line of 
the rhinoceros is pecked out in a dark silhouette. This example illustrates 
once again how misleading it can be to classify works of art merely from 
a technical point of view, without giving any consideration to their deeper 
meaning. Seen in isolation, this monochrome silhouette might be thought 
to represent an early stage of evolution; but in fact we are dealing here 
with a more advanced, animated and fluent style than that of the large 
static rhinoceros at Leeufontein; since in point of subject-matter it 
probably corresponds rather to the small and apparently older picture 
under its belly, these two silhouettes are related to one another. 
In the plate on p. 155, however, we have a work in which the inner surface 
is 'stopped out' to form narrow ridges and which, taken as a whole, shows a 
high degree of psychological empathy. The dark silhouette is explained 
by the nature of the theme, which is darkness — as at Lascaux, pigment is 
sprayed by means of a blow-pipe on to certain parts only of the rhinoceros 
and the bison's belly. The impression evoked is as though the guilty demon 
has been caught fast in his own snares and struck with blindness. 
These figures therefore express an ethic which is very far removed from 
the conventional idea that they served as a utilitarian means of obtaining 
prey through magic. To attribute such a purpose to rock art shows the 
lack of imagination characteristic of the modern world. 
MOTIVATION This latter remark automatically raises the question of the motivation 
underlying South African rock art. 

In discussing engravings we have mentioned cases where the artist's 
imagination was easily stimulated by slight similarities of form and 
structure, and even by natural vesicles in the rock, which sufficed to 



184 



evoke a certain image in his mind — just as the granite cave at Nswatugi 
inspired him with a vision of the cosmos. But these stimuli, which led to 
the creation of ' cosmomorphic' pictures filled with emotional intensity, 
were merely external; the real motivations lay deeper. The artist's mind 
was overflowing with these images, as it were, and in the rock before 
him, in the perfect vaulting of the cave, he found the revelation of his 
earth-bound perceptions. Thus the pictures of these animals, which he 
'sought with his spirit', so to speak, are filled with intrinsic significance. 
With his lively instinctive sense, he was ever on the look-out for manifest- 
ations of his mythological beliefs. At the same time he kept an alert eye 
upon the terrestrial world, for the heavens appeared to him to abound 
in figures suggestive of the splendour and beauty of the animal kingdom. 



Below, innumerable superposed paintings in various colours and sizes: above, a number of kudus, 
over which towers a group of giraffes and zebras. Between these and the uppermost group of 
giraffes (reproduced on front of slip-case) are scattered small human beings in dark red. Main 
part of rear wall at Nswatugi, Matopos, Southern Rhodesia. 




; ^y 



r w 




185 



It might almost be considered miraculous that these animal pictures, for 
all their mythological significance, are nevertheless so naturalistic and 
down-to-earth. But could we really expect it to have been otherwise? The 
man who kills an animal for his own subsistence is most likely to feel a 
sense of awe when beholding its striking beauty; how much more must 
this animal have meant to him when he thought it a divine being! 
Drama of Nevertheless, the decisive factor is still the part played by each animal 
cosmic change j n tne g reat (J rama f cosmic change. 

The esoteric ethical value with which each individual animal was endow- 
ed by virtue of its place in a higher world was made clear to the young 
hunters when they were being initiated into mysteries. There can be no 
doubt that these pictures also played a part in the propaedeutics of 
religious beliefs. This is evident from the mere fact of their relation to 
mythology and the naive way in which information is imparted in the 
form of fables. Moreover, we know of actual examples of propaedeutic 
exemplars and precepts taught to young Bushmen. These often contain 
straightforward explanations of the reasons why certain animals have a 
particular character in mythology. "Don't you see that the head of the 
cow antelope resembles Mantis? That is why he loves this animal." This, 
we are told, is how their fathers taught them the meaning of these fables. 
Animal parable The plate on p. 189 may serve as an instructive example of the way in 
which the Bushmen interpreted their mythological identification of an 
animal (cow antelope) with an insect (praying mantis) and at the same 
time also with the sun. In Fig. 49 we have a similar representation of the 
myth about the moon and the eland. The same disproportion between 
illuminated head and sketchy body is also met with here. But where the 
eland was a static reposed figure, in this case the cow antelope is depicted 
in motion — like the hastening sun; it is regarded as the swiftest animal, 
and it is probably for this reason that it is related to the sun. The 'resem- 
blance' to the mantis is not merely based on its curved horns, resembling 
the posture of the raptorial limbs of this insect, but also on the fact that 
when the antelope gallops its forelegs are bent, thus forming another 
curve. 

It might be thought that the body has been crudely touched up by some 
amateur and that it is much more recent. But the very fact that only two 
legs are rendered shows that it dates from a particularly early phase, as 
is also apparently indicated by the twisted perspective of the horns. This 
mode of representation thus had an esoteric motivation: it was designed 
to convey a mythological idea to the initiate — as is clear from the 
remark made by the young Bushman quoted above. 

186 




Giraffe and quagga without spots or stripes, over a jumble of smaller animals painted in a darker 
colour. Detail from rear wall (Plate p. 185) of the granite cave at Nswatugi, Matopos, Southern 
Rhodesia. Height of giraffe 5 ft. 3 in. 



But at the root lies the idea of the sun regenerating itself, in that, accord- 
ing to the myth, as its first rays appear over the horizon the severed 
limbs of the animal are again joined to its head of their own accord. 
As in all ancient cultures in which these deified animal figures were 
worshipped, we have images of these gods (idols). The believer was no 
less close to the person of the deity where the latter was represented in 
animal form. Thus the history of this art comes into its own only when 
it is seen as an attempt to grapple with sublime religious ideas, with 
a spirituality and transcendentalism that cannot seem to anyone today 
so very strange and remote, since behind these animal figures there stands, 
invisible, man himself with his ideas of the supernatural. 
Man is a humble creature, inferior in rank to his fellow-animals; in them 
the enigma of life is manifested in such a pure and simple way that he 



Idols 



187 



need only comprehend this to recognize that he is a human being. This 
is how it is explained in the fables: in animals one can see human beings; 
they are merely the various masks donned by men; for man is a perpetual 
initiate, ever marvelling at his own existence; were he to divorce himself 
from his animal masks, he would merely be a wretched creature, driven 
from the beneficent universe. 

In this primeval age the world, of which man with all his human faculties 
formed a part, was still whole and undifferentiated; later man could only 
revert to this primeval unity orgiastically, or — much later — emotionally 
and romantically. 

Modern man approaches this art from a purely historical point of view. 
In general he attempts to find his bearings in it by expecting it to provide 
confirmation of his own theories of evolution. 

There is indeed no better opportunity to free oneself from the restrictive 
fetters of modern intellectualism than by endeavouring to comprehend 
South African art. 
Art is unhistoHcal This art is not only ageless, but in the narrower sense is unhistorical. If 
we do not recognize this fact we are continually in danger of falling into 
error or at least of overlooking essential points. For most of us this art 
can tell us little about ourselves. The second approach towards 'clas- 
sifying' this art, the topographical and geographical one, also leads to 
continual disappointments, since it adheres in equal measure to super- 
ficialities, such as ethnographical and geographical boundaries. 
We have hitherto deliberately made no distinction between two different 
categories of rock art which scholars maintain should be treated separately, 
not only because their essential characteristics are different, but also 
because they were produced by completely different cultures and are, 
moreover, not contemporaneous — not to mention the deeply-rooted 
prejudice that all art in South Africa must be much more recent than 
that of Europel Instead of going into this controversy in detail, we shall 
adhere to that never-failing source, mythology, which throws adequate 
light on the subject. What is really important, in our view, is to com- 
prehend art from the standpoint of its spiritual content, its matrix — 
particularly when, as in this instance, we have the unique opportunity 
of learning about it from the living tradition of the people who 
produced it. 

In the plate on p. 160 we have a fine engraving of a quagga — a sub- 

Significance for species of zebra that became extinct only a short while ago — on a fragment 

history of art Q £ diabase in which particles of molten quartz are still to be seen. Because 

of its significance for the history of art this rock, together with some other 



pieces, was taken by the author from its site in the western Transvaal 
to the collection at Pretoria University. The original is not much larger 
than this reproduction. And yet a great deal may be deduced from it Plate p. 160 
about the history of South African art. 

At first sight it looks as though we have here nothing more than a prelim- 
inary sketch executed in a fine engraving technique, with part of it in 
pecking technique. This is a delusion. First we notice that a natural 
crack in the rock, not far in front of the quagga's mouth, gave the original 
suggestion of a horse's mouth, just as similar natural details have been 
utilized on other occasions by rock artists in Europe as well as in Africa. 
In the first version this natural line was probably simply continued to 




Cow antelope running. The entire head is executed in silhouette, but part of the stunted body is 
only represented in outline; this is connected with ideas of the cosmos. Maretjiesfontein, western 
Transvaal. Length approx. 93/4 in. 



Suggestion of 
pictures 



Phases 



Engraving and 
line engraving 



produce the prototype of the present head. The additional lines which 
must once have existed seem to have fallen victim long ago to weathering, 
despite the hardness of the rock. 

This earliest outline of the head was probably bounded at the lower 
jaw by a slight ridge in the rock, so that it stood out in relief. This would 
correspond to the plastic conception of our large rhinoceros, for example. 
The head was thus larger and of a heavier type than the present elegant 
head of the quagga. Perhaps this was even a rendering of a more heavily- 
built species of horse, of which there are many varieties among those 
long ago extinct. In any case this head was superseded by another, still 
just visible, of which the nose commences approximately half-way between 
the old crack and the most recent head. The line runs roughly parallel 
to the present profile and y 4 in. from it; but it was adapted still more 
exactly to the original profile and had a greater curvature, resembling 
that of the equus capensis and other old species of horse in South Africa. 
In the second engraving one can also still easily pick out in its proper 
place the ear turned towards the front. 

Even the present outline of the quagga's head and neck, which is the 
first to have the inner surface filled in, is not the next version, but is 
clearly a more vigorous retouching. The lines are more deeply incised 
and more defined than those of the rest of the body, of which in our 
plate we can only see the lines of the back, chest and belly, running diag- 
onally. We accordingly have a sequence of four versions executed in the 
engraving technique, between each of which there was an interval long 
enough for the engraving to become completely invisible in the hard 
rock. And since this phase (which, as the most simple outline drawing, 
corresponds to the Aurignacian in Europe) nevertheless was preserved 
in some places — as it was in this instance — this enables us to realize how 
durable such engravings are. 

But this was not yet the end of the evolution of this quagga picture; much 
later the engraving was retouched in the new pecking technique. From 
the lighter patination it can be distinctly seen that a considerable length 
of time must have elapsed between the touched-up engraved lines and 
the pecked-out grooves, which still have a bluish shimmer about them. 
The evolution illustrated here is by no means a special case, but can 
be substantiated everywhere. The art of line engraving passes through 
various stages: first the simplest outlines with only one foreleg and hind 
leg shown; then a crude representation, gradually coming closer to nature 
and to true perspective; then complete mastery of the art of filling in 
the inner surface and portraying movement; finally, with the pecking 



190 



technique, the addition of colour and the development of the artistic 
vision of a' real painter. 

The final product of our unfinished small quagga figure would take us 
approximately to the 'stage' of the large rhinoceros and the giraffe 
discussed above. But how are we to date this stage? Two stages already 
existed: and one must rule out any question of a sketch, or of any in- 
fluence upon these two stages by the artist who did the pecking. Perhaps 
each stage was separated from the other by centuries, and probably even 
by millennia. 
As a result of this examination of its genesis it can be established that 




F'g- 53 — R°> n myth, as in Fig 
H. Obermaier and L. Frobenius 



50, from Alt. Ain Gudeja, North Africa. After 



South African rock art spurns any kind of historicism. It is, like mythol- 
ogy, ageless; a picture commenced thousands of years ago can be taken 
a stage further by artists in our own day. An effort could, it is true, be 
made to take technique as a criterion, and to establish (in this case, at 
any rate) that the pure form of line engraving represents an older stage. 
It is, however, certain that, once pecking came to play a dominant part, 
it ceased to be employed as the sole means of representing the entire 
outline of the figures. On the other hand, there is no evidence that pecking 
did not appear already at the time of line engraving; indeed, everything 
points to the fact that it is at least just as old, or even older. 



Technique as 
criterion 



J9 1 



In Europe, as throughout the whole of Africa, we find broad grooves 
as outlines already from the very beginnings of rock art. They originated 
with drawings made with the finger in sand or clay and are the primeval 
form whereby a simple rite performed at an open hunting-ground was 
represented on stone. As Leo Frobenius established in the case of the 

Primitive cult Pygmies, this primitive cult has survived right up to the present day. 
Our works of art are thus deliberate acts of immortalization, which derive 
their origin directly from a ritual act, from a technical point of view 
as well. 

What is the source of this fine art of engraving, of which the origins, 
in South Africa as elsewhere, go back to the remotest past? Engraving 
is in essence illustrative, and the fact that in Europe it was at first 
applied to small movable objects, such as tablets and bones, and only 

Propaedeutics later to walls, seems to point to its use in propaedeutics, i.e. in the 
expounding of myths. The same applies, of course, to South Africa, 
although we do not find here any movable objects decorated in this way; 
instead we occasionally come across miniature drawings incised upon 
largish or smallish stones conveniently located on exposed hills bearing 
pictures. In many places there are hundreds of these drawings, frequently 
superposed upon one another, and at the older sites they are sometimes 
not pecked. 

In their origin the broad-grooved outlines must derive from actual cults, 
whereas the fine engravings must derive from the instruction given to 
those being initiated into mysteries. In the course of time the two forms 
will have been blended, or even have superseded one another, as is the 
case with our quagga picture. In Europe, too, the same combination of 
engraving with painting occurred, so that the two techniques have thus 
constantly fertilized each other. Painting may have exerted an influence 
upon the abundant pecking out of the surface, so that a colour effect 
was obtained; but such a colour effect could also easily have been produced 
simply by working the stone. 
Colour For corroboration of this opinion, we may consider the eland in the 
plate on p. 193, which has a magnificent colour effect. In this engraving, 
which is in the very important collection in the Transvaal Museum, one 
can distinctly see the colouring of the rock: the core is bright blue and 
the layer formed by oxidization a brownish purple. The animal, which 
is of splendid physique, is modelled with the utmost sensitivity, even the 
'texture' of its coat being beautifully brought out; it is not incised deeply 
into the blue core but is only cut through the dark surface layer as far as 
the core, thereby obtaining a silvery tone of colour. Particularly exquisite, 



192 




T*S 




Eland buck on diabase, in combined engraving and pecking technique, with the inner surface 
'stopped out'. Western Transvaal. Transvaal Museum, Pretoria. Length of animal approx. i ft. 
j3/ 4 in. 

both as a painting and as an engraving, is the light pecking in stripes to 
indicate the mane; even the hair on the forehead is depicted (this is an 
exceptional feature) by means of delicate engraving. In this respect this 
work reaches an acme of perfection in the art of engraving. But certain 
signs point to the existence of an early prototype, or at any rate to a stage 
when this art had not yet been completely mastered: the horns are ren- 
dered in twisted perspective, as in the plate on p. 189; that is to say, the 
horn which in foreshortening is normally visible in front of the other is 
in this case behind it. The same applies to the forelegs: the one further 
away in perspective appears to be longer than the one closer to the 
observer. 



193 



This last point must, however, be treated with caution. We have already 
referred at the beginning to the fact that in mythology the growth of the 
animal embodying the moon is suggested by the fact that the legs often 
still appear to be stunted, and the plate on p. 172 shows the same phenom- 
enon as can be seen in Fig. 48. It has even been established that as a rule 
the legs of almost all the eland figures in the Drakensberg region appear 
too short, and this has even been regarded as a stylistic feature distin- 
guishing this whole group of works! 

In any case the elaborate care with which the coat of this animal is 
treated is often clearly explained in propaedeutic tradition. The legend 
of the creation of the elands — by the sun (Mantis) out of a shoe placed 
in the water — emphasizes the point that Mantis anoints the coat of his 
little favourite with liquid honey to give it a beautiful sheen. What the 
artist particularly sought to accentuate by his unusually elaborate execu- 
tion was the analogy of the silvery moon waxing to fulness; and no image 
can convey this so beautifully as that of this exquisite animal, with its 
velvety coat of honey, quietly browsing. 
Rock galleries When we go on to consider the paintings of the Drakensberg rock gal- 
leries, in which elands are the prevailing theme, we have here veritable 
masterpieces of composition in colour, inspired by real poetic imagination. 
When the moon rises the hunter gives expression to his joy by ritual 
dancing and singing, which go on all night long; no celestial body is so 
close to his heart as the moon; to women it is a mysterious goddess, to 
men it is an auspicious sign of clear nights for hunting — this recalls 
the part played by Artemis in Greek mythology. 

The Bushmen of the mountain caves had a most revealing myth, handed 
down by Orpen, which shows the transcendental significance these 
pictures possessed. The following tale is told about one of their deities: 
"Quanciqutschaa killed an eland, cleansed himself and his spouse, and 
ordered her to grind some canna. And he sprinkled the ground with the 
powder, and all the elands he had killed rose up, and some came with 
javelins sticking in their bodies which had come from those who wanted 
to kill him. And he pulled out the javelins, a whole sheaf of them, and 
they stayed with him; it was a place surrounded by hills and deep 
ravines. . ." 

This is a distinct reference to one of the typical mountain picture galleries. 
In the lighter sandstone of the mountain massifs which, where it is suspend- 
ed beneath firmer rock strata, weathers and falls down as large boulders, 
galleries were formed that were protected by a kind of projecting roof. 
The artists here found a wall which provided a tempting surface for 

194 



painting, in colours that became ever brighter in the course of time. A 
popular solution of the problem is a frieze extending along the rock 
ceiling. Could there have been anything closer to the hunter artist's heart 
than to paint in the form of animated figures the cosmic myths in which 
he believed, and the joy he felt at the earthly incarnations of the celestial 
bodies? 

The plate on p. 163 shows, in addition to smallish red animals and small 
human figures which correspond to them in colour, size, and lively 
vigour, a file of reddish-white elands darting along, and also, a little 
higher up, some cow antelopes — the fastest animals on earth, which in 
mythology embody the sun (Mantis). In this composition we thus find 
again the same pattern of cosmic imagery as in the Rhodesian cave: 
inferior men and animals are related — indeed, are interchangeable — 
as required by the metabolic conception of life; great celestial luminaries 
rise into the supernatural sphere, lending enhanced significance, colour 
and grandeur. 

In such galleries dances and rites were often performed far below this 
frieze, but in the lower galleries they were frequently performed 
opposite it. Among the animal figures paintings of men were added, 
dancing and leaping wildly to express their unbounded joy at being 
alive and at the promise of abundant game which each new moon 
denoted. In each case he gives figurative shape to his emotions by the 
performance of a solemn rite, by dancing, or by the joyous pursuit of 
animals which do not even appear commonly as prey; for he is here 
giving expression to his inner tension. And plump women are also 
frequently portrayed to denote the same feeling of exuberance at the 
blessings of life. 

Sometimes hymns of praise to the gifts bestowed upon man by the moon 
are characterized by a greater stateliness and sense of repose, as when 
elands are shown congregating in large groups (cf. Plate on p. 197). In- 
exhaustible is the hunters' joy when contemplating the impressive size 
and ample proportions of these magnificent animals, with their bodies 
glittering like gold; the artist represents them resting, browsing, seen 
from all angles, and even daringly foreshortened; it is as though he is 
inviting the sun, the source of all fertility, to accept with pleasure the 
sacrifice offered to him. 

Such idyllic animal scenes suggest an Elysian existence in the midst of 
nature — an impression enhanced when, after examining these pictures, 
we come across one of the herds of these fine animals in their natural 
surroundings in the Drakensberg reserve. 



195 



Motifs But closer study of the picture soon reveals motifs of undeniable myth- 
ological import. First of all we may point to the motif of recumbent 
animals, particularly the cow at the bottom on the left-hand side, which 
is being approached in a playful manner by an ox. There is a striking 
parallel with the famous pair of reindeer from the cave of Font-de- 
Gaume in the Dordogne. The fact that this picture represents the myth- 
ological concept of the male animal rousing and cherishing the female 
moon is distinctly emphasized by the stunted legs of the female reindeer. 
Ritual dances But the connections with ritual dances go much deeper still. A young 
Bushman girl is isolated at the time of her first menstruation and initiated 
by an old woman into the mysteries of womanhood. When she is once again 
permitted to show herself to her fellow-tribesmen a dance of elands 
is performed in her honour. Two fairly old men disguise themselves 
as bull elands, and the girl lies down in the middle of a circle of married 
women, who dance around her and encourage the 'bulls' by exposing 
their buttocks and making provocative gestures. In this way the girl 
is given an illustration not only of the mystical relationship that hence- 
forth will exist between her and the moon, but of the whole myth of 
the fertilization of the moon by the sun. 

It lies beyond our scope here to discuss whether the famous figures of 
Foreshortening, the cow bisons depicted on the cave roof at Altamira are connected with this 
ritual dance, as we believe to be the case; but we must certainly make 
the point that a connection exists with our foreshortened rendering of 
the cow eland. There is no explanation whatsoever for the introduction 
of foreshortening other than as an indication of this principal motif in 
the dance of elands performed in honour of young women. And this 
leads us once again to warn against laying down a preconceived 'evolution' 
for art! Foreshortening is not an achievement reached only at a certain 
stage of development, as one may be inclined to think. If a motif with 
a profounder meaning is to be represented in visual form, some artistic 
means will always be found of doing this, as is shown here already in 
prehistoric art. This again seems to show what a decisive influence 
mythology and ritual had upon this art. 

When one surveys these delightful pictures of elands, executed in such 
varied techniques and colouring, and spread over so vast an area, one 
asks oneself the question whether this diffusion can be solely responsible 
for these very considerable differences in style, and whether they were 
not produced by some chronological development, even at the same 
locality. If one looks beyond the boundaries of the Union of South 
Africa to Rhodesia and further north, eland figures no longer take up 

196 



the almost exclusive position in the caves which they occupy throughout 
the South African subcontinent, and again the question arises as to the 
connections between these different regions, and the stage of artistic 
development to which each figure belongs. 

In the case of the engraving shown on p. 193, one would be inclined, 
despite the 'unique' southern engraving technique, to think of European 
parallels from the flourishing Magdalenian period; in the case of the 
fleeing elands shown on p. 163 (especially if one includes the small 
red male figures) Spanish Levantine art comes more readily to mind. 
The larger composition on p. 197 is reminiscent of the roof at Altamira, 
and not merely on account of its motif. On the other hand, the elegant 
sketch-like character of the little picture from the eastern Transvaal 
(Plate on p. 172) brings to mind Mediterranean rock art, such as that 
of Levanzo. 

But it is impossible to divide up this compact group of southern paint- 
ings and to assume that any influences could have been felt in particular 



Group of elands resting. Above, on the left, an animal in a faded dark colour; all the odiers are 
bichrome. Bold postures and foreshortening. Near Bethlehem in the Drakensberg Mts., eastern 
Orange Free State. Length of individual animals approx. /_?% *'"• 





197 



localities, while much older phases of European art survived in neigh- 
bouring districts. The compactness of this southern group, evident from 
the prevalence of elands as a theme, points to the fact that it is an organic 
entity and has a common mythological basis. 

Let us now attempt to substantiate this in greater detail, at least as 
far as the southern region of rock art is concerned, by going back to the 
engraving of the quagga as a link between the group of engravings in 
the interior of the country and the cave paintings of Rhodesia. For these 
groups are thought of as separable from each other. The northern group, 
comprising Rhodesia, the northern Transvaal and South-West Africa, 
is divorced with some justification from the two southern groups in 
the remaining part of the subcontinent: that in the Drakensberg range 
and that in Cape Province. 

There are of course also different dialects among the Bushmen, who 
were once diffused over the whole country; and undoubtedly local 
myths developed as well, for these hunters had been settled here since 
time immemorial. But the substance of all the legends is a single indivis- 
ible primeval myth closely linked to beliefs, rites and cults. 
This fact is also corroborated by archaeological evidence. We have 
already mentioned that at a certain stage in the South African Middle 
Stone Age (i.e. in the Middle Palaeolithic) all the older industries, even 
divergent ones, reached a climax in the Magosian, which is to be found 
distributed evenly over almost the entire continent. This phenomenon 
Uniform culture may be regarded as evidence of the creation of a uniform culture and 
system of beliefs, which also found expression in tool-making. It denotes 
the appearance of a new intelligent type of man. 

In parts of the great double-arched cave of Pomongwe in the Matopos, 
south of Bulawayo in Southern Rhodesia, there are some ancient outline 
drawings in a dark colour. Remains of such figures of animals, some o r 
them drawn on a large scale, are to be found everywhere in this northern 
group; they are the oldest remains extant. The plate on p. 174 shows, 
in addition to some unidentifiable outlines, a fine drawing of a zebra 
surrounded by partially effaced silhouettes in purple and a fine bichrome 
painting representing a creature flying and holding in its hands a bundle 
of switches or spears. To the left and right of it are some human figures 
in the same purple colour as the remnants of the animals. 
It seems as though this insect-like flying creature is illuminated from 
below. Unfortunately it can no longer be established with certainty what 
it represented; but there is a good deal of evidence that points to the 
fable of the cow antelope being dismembered and resurrected, where 

198 



Mantis also appears as an 'old wood-carter' who strikes terror into the 
hearts of young girls. As he dances through the air, we are told, the 
soles of his feet shine: there are of course allusions throughout to the 
manifestations of this Bushman deity as the sun (cf. p. 194—5). 

We are, however, mainly interested in the appearance of a line drawing Line drawing 
among paintings as well as engravings, which clearly shows that engraving 
is the oldest branch of rock art. It can unfortunately no longer be estab- 
lished whether there was originally a second colour which denoted the 
light and dark stripes. This is unlikely, for wherever else such bichrome 
zebra figures occur in this northern group, a preliminary sketch of this 
nature does not exist, and the dark stripes were applied directly on to 
the white ground-colour with broad sweeping strokes. 
If this supposition is correct, then this would be a stage of painting 
coinciding with pure line engraving. That the stripes were 'filled in' 
later is certain; this method of applying stripes would therefore correspond 
to the more recent one already mentioned. We also know that there is 
a whole group of these engravings devoid of any 'painted' pecking. The 
cardinal point is that painting had some part at any rate in this graphic 
phase; for the surviving remains at the cave of Pomongwe show the same 
evolution (from an older drawing in outline, without any filling-in of the 
inner surface, to a more advanced stage with such filling-in), such as has 
been observed in the case of the quagga figure reproduced on p. 160. 
The dorsal outlines of the animals to the right and left of our zebra, and 
below it, indicate such stages of stylistic evolution; and in this case we 
need have no hesitation in establishing a homogeneous development run- 
ning parallel in every respect to that in Europe. We find exactly the same 
pattern: from the oldest foil-like drawings with only two legs visible 
(the Aurignacian stage) right up to the most accomplished representations 
of the most advanced Magdalenian, with the meticulous rendering of 
detail. 

Finally, mention must be made of another zebra figure without any 
stripes: it is to be found in the cave at Nswatugi already discussed, on the 
right next to the large giraffe, and is executed in the same style. The Plate p. 187 
simplest explanation of the problem would be to draw a comparison 
with the technique of this giraffe painting, but since we have no Bushman 
legend relating to the zebra, we have to fall back upon a curious Hotten- 
tot tradition. According to this the zebra is said to have originally had The zebra's stripes 
no stripes at all, until it found the sun by the wayside and put it on its 
back. But the sun refused to be taken for a ride in this fashion, scorched the 
zebra and cursed it, and then seated itself on the horns of the bull instead. 



Homogeneous 
development 



199 



The first part of this legend is probably Bushman in origin; the second 
part indicates the transition from hunting to cattle-raising, with the 
emergence of a tabu as regards horses, and probably horse-meat as well. 
This process, important from the point of view of cultural history, is 
represented in many North African rock pictures, in which we can note 
renderings of the sun in its manifestation as an ox. (As is well known, and 
need not be elucidated further here, the Egyptian cow goddess Hathor 
belongs to the same tradition). 

In any case this legend provides a clue to the divorce of the Hottentot 
pastoralists (with their admixture of elements of Hamitic and North 
African culture) from the Bushman hunters, who are in some respects 
related to them. 

Let us now consider briefly whether our figure of the zebra without any 
stripes is in keeping with this legend, and whether this is an allusion to 
a mythical primitive state — as in the case of the giraffe 'without any spots'. 
For it is quite possible that the legend was inspired by the fact that the 
quagga, a kindred species, which was a more favoured subject in rock 
art (and for this reason must also have had a higher mythological rank) only 
possessed stripes on the front part of its body and was almost completely 
brown on its hind quarters, like the whole of this figure on the cave wall. 
At first we are struck by the fact that the animals appear so close together 
and are almost entangled with one another. The brightness of the picture 
has already been stressed above. The fact that both figures are the work 
of the same artist indicates that they belong to a single group, that of 
renderings of a subject referred to in Hottentot legend as 'riding'. This 
could also suggest fertilization; for this is probably the deeper meaning 
of the connection between the sun and the moon represented as a bull. 
What then would be the significance of the giraffe in this connection? 
Fig. 50 Let us now go back to the picture of the elephant at Philipp Cave: the 
head of a giraffe has been painted over the belly of a cow elephant, as 
though penetrating into it and trying to reach the little springbok inside. 
Might it be possible that in this case the giraffe, too, is an embodiment of 
Mantis? This is suggested by the explanation given to Bleek by a wise 
old Bushman: "Don't you see that Mantis looks like a female antelope? 
This is why he loves this animal." It was in this way that former genera- 
tions must have explained to young initiates the relationship between 
Mantis and the antelope. 
Assimilation In the case of the cow antelope this 'resemblance' is based on the curvature 
of the horns, which corresponds exactly to the posture of the raptorial 
limbs of the praying mantis, just as 'consanguinity' was also suggested, to 



life- i 


















T*-*H% 



H '■>. 






■ v- : 







:V "•*-:;- i 



^, 



k?f«S 




Two isolated kudu cows, painted over small blurred figures of earlier date. Destruction by missiles. 
Left-hand side of rear wall, Nswatugi, Southern Rhodesia. Length of animal approx. i ft. 4 in. 



a keenly observant eye, by the way in which both the insect and the 
animal have their heads turned sideways. The term 'sympathetic magic' 
has been coined to denote the drawing of parallels in this way by prim- 
itive peoples between things which according to our conceptions are 
completely unrelated. But (so far as the Bushman is concerned, at any 
rate) this overlooks one fundamental fact: the metabolism of a pan- 
theistic Weltanschauung, according to which insect, animal and man are 
seen as consisting of identical 'substance' and as identical manifestations 
of a supernatural Creator. 

Anyone who considers the giraffe from this point of view will be quick 
to see the parallel, particularly if one observes this animal whilst it is 
drinking: in this posture its forelegs are bent, the rest of the body is 
held erect, and when it looks at something it turns its diminutive head 



Spiritual aspects 



The Bushman 
as teacher 



Prntohisloricai 
elements 



sharply to one side; this suggests most convincingly that there could be 
some intrinsic relationship between the giraffe and the praying mantis. 
Since Mantis embodies the sun, its metabolic animal figure, the giraffe, 
may be regarded as substantiating further our interpretation of this 
legend. 

This also provides a ready explanation of the way in which both the 
giraffe and the accompanying zebra are represented. The zebra is shown 
at the moment before it puts the sun on its back and is 'scorched' by its 
rays. Indeed, this even explains why this group was painted at this partic- 
ular spot 'on the horizon', below the large giraffes: in order to put the 
sun on its back, the zebra must have found it close to the ground, i.e. on 
the horizon. The punishment inflicted on the zebra is the natural con- 
sequence of its presumption in failing to recognize that the giraffe is a 
superior creature. The same idea is expressed very well in the myth of 
the cow antelope being dismembered by the young girls: after the failure 
of their attempt to dismember a spiritual being, they do not simply go 
home in disappointment, but are stricken with terror, and on their return 
home are taught a stern lesson by their father. 

Such an admonition may not be entirely out of place with regard to 
students of rock art: before one draws conclusions about 'style', it is 
advisable to become familiar with the higher purpose by which the artist 
was motivated. If we wish to understand the spiritual character of this 
art, we must beware of approaching it with our concepts of evolution. 
There is no inconsistency between the giraffe represented as the sun 
and as a star, for the eland, too, can be a constellation. Perhaps this 
twofold meaning may actually explain the duplication of the figures in the 
cave at Nswatugi, and enable us to understand more clearly their com- 
prehensive cosmic-terrestrial structure. 

The blending of legendary motifs through the contact between the pas- 
toralists and the older hunters can be followed in rock art. Although 
the pictures in the Drakensberg show this art in its death-throes, even the 
Bantu recognized the maxim: "the Bushman was our teacher"! 
In Rhodesia there seems to have been an assimilation with other proto- 
historical cultures. In the painted caves of this group extraneous influences 
can be clearly identified. 

A taller type of man appears; there are traces of new cults; men are 
portrayed in action; and they carry javelins and heavy loads upon their 
heads — as do immigrants or nomads. New types of dwelling spring up, 
and these, too, are depicted. In the south we even find occasional render- 
ings of herds of cattle. In some localities the caves contain granaries, 



like small huts, built of clay and sticks. These huts are even painted on 
the walls, indicating cultivation of the soil and rites associated with it. 
The whole'north-eastern part of the country long inhabited by hunters 
comes to be dominated by the mines constructed there; mysterious stone 
buildings (Zimbabwe) tower up massively on the slopes of the granite 
hills. Tin, copper and gold, and even iron (!) are smelted on rocky perches 
and in caves, as though in the great cauldrons of primitive alchemists. And 
all this is shrouded in mysterious protohistorical darkness. 
The plate on p. 169 may illustrate this, the final battle for survival 
fought by the ancient hunter culture. This is part of a wall in the impres- 
sive cave at Silozwane, situated high up on a mountain side, and reached 
by a precipitous climb over bare granite rocks. The whole vault right 
up to its roof is dominated by naked human figures, larger than life, 
painted in dark red and yellow. From them there emanates a distinct 
aura of solemnity, as in a sepulchral vault, and the animals of the hunters 
seem crushed and cramped. On the lower section is a jumble of little 
sacred pictures, the work of many millennia, which has been painted 
over with a dark colour. Across it stretch gigantic snakes, extending almost 
round the whole circumference of the cave. A new era has dawned, and 
the ancient sacred chamber is now in the hands of a new race of men. 
The animals representing stars, the giraffes, are overrun by large human 
figures. 

One senses the atmosphere of a later era, a time of ancestor worship, ritual 
sacrifice and sacral kingship. 

The delightful and graceful little pictures of kudu from Nswatugi, in 
the ancient granite district of the Matopos, reproduced on p. 201, are 
two final gems, in which the vanished world of the hunters is depicted 
in very light colours on the glittering blue vault of the cave. As though 
able to sense the impending calamity, the timid animals part company 
and prick their ears. Never before has the almost divine purity and 
beauty of the animal kingdom been painted with such tranquil calm, and 
at the same time with such emotional depth. 

The shadows of animals and men from bygone ages hover about their gold- 
en bodies. Here the most recent and accomplished stage of hunter art has 
been reached, expressive of serenity and profound spiritual feeling. 
It has been suggested that the damage done to them resulted from the 
vandalism of European big-game hunters, who were tempted to aim 
their fire-arms at paintings of animals that were so true to life. This sounds 
like a joke in bad taste. It is more probable that, as in other prehistoric 
caves, the picture was destroyed in the course of some ancient ritual. 



203 



THE ROCK ART OF AUSTRALIA 

BY 

ANDREAS LOMMEL 



The best point at which to begin the study of Australian rock art is with 
the so-called wondjina figures of north-western Australia. 
The wondjina paintings are to be found beneath overhanging rocks 
affording protection against the tropical rains. Such rock surfaces fre- 
quently occur on the large isolated stones that rise from the flat tableland; 
these are often mushroom-shaped, i.e. on a comparatively narrow base 
the rock extends outwards on all sides, so that all the pictures beneath 
the overhang are protected from the rain. Occasionally painted rocks 
are also found on ridges which can only be reached by a difficult climb. 
Still others are situated in deep ravines or in rock crevices running 
parallel to a ravine; the latter are particularly hard to find. Unless one 
is accompanied by a native familiar with the district, one can easily spend 
days searching the sides of a ravine only about two to three miles long 
before one finds the painting. Other pictures, by contrast, are visible from 
afar; already when one is half a mile away one can see that the rock 
is painted, and as one rides up closer one can gradually make out the 
figures. 

The wondjina figures are anthropomorphous representations, generally 
ungainly and often crudely worked. Sometimes, however, they have a 
certain primitive ingenuity, and very often the natural protuberances 
and recesses of the rock are utilized to give the effect of relief. A wondjina 
is usually rendered in a recumbent posture. The face is enclosed within 
a broad horseshoe-shaped band in red or yellow ochre. Only the eyes and 
nose are depicted, not the mouth. There are several explanations from 
mythology for the absence of the mouth, but since they do not agree 
one feels inclined to call all of them in question. The body of the wondjina 
is generally painted in shades of white and filled in with vertical stripes; 
the arms and legs are sharply defined; the hands and feet are in most 
cases very rudimentary; and the feet are often rendered in such a way 
as to show the sole, i.e. instead of the foot we have a footprint. This 
is not really surprising if one considers the important part played by 
tracking in the lives of these hunters, and if one bears in mind that 



205 



throughout Australia footprints are often depicted in lieu of the creatures 
who made them. The wondjinas wear on the chest a longish object 
which may denote either the heart or the breastbone, but this explanation 
is not convincing. The sex of the figures is only very rarely indicated, but 
despite this the aborigines are mostly very certain in stating whether a 
wondjinas is male or female. 1 

Next to the wondjina figures, above them and on the top of them, smaller 
wondjina figures are often painted, or merely suggested by their heads. 
These are 'children of the wondjinas or the souls of men who have 
developed out of this wondjina. There are also numerous representations 
of animals such as kangaroos, fish, birds, dingo dogs and opossums, as 
well as edible plants and bulbs, showing that these living things developed 
with this wondjina in primeval times and that their souls have a habitat 
in this picture. Sometimes inexplicable longish figures are also to be 
found in the pictures, which the aborigines refer to as honey; these have 
a purely symbolic character. Honey is a sacred food; and certain painted 
ritual sticks are said to serve the purpose of helping to find honey. The 
representations of honey are in fact renderings of these ritual sticks. 
Close to the wondjina figures, in smaller niches, there frequently occur 
rather grotesque human figures called by the aborigines 'demons'. These 
are sometimes painted over the wondjina. 

A wondjina picture ought to be painted afresh every year by the chief 
of the tribe appertaining to it. Even nowadays one comes across freshly 
painted pictures, although only in rare instances. From those that are not 
freshly painted it can be seen that they fade and flake comparatively rapid- 
ly, since the red and yellow mineral pigments consist of ochre, the white 



Fig. 54 — Fish. Painting in red in so-called 
X-ray style. Oenpelli, Unbalania, Northern 
Australia 



206 




^r^l^@: ... 




Recumbent wondjina figure with plum-tree. Wonalirri, North-western Australia. 19 ft. 8 in. X 
4 ft. 11 in. 



pigment of white clay, and the black pigment of ground charcoal; and 
these are applied after being mixed with water alone. At the site of the 
picture one often still finds remains of chewed sticks that served as 
brushes, sheets of bark that served as palettes, and pieces of charcoal. When 
painting the figures afresh, the artists always keep to the same subject, 
but not necessarily to the lines drawn by their predecessors. Thus under 
a freshly-painted or relatively recent wondjina picture one can often 
easily spot the contours of an older figure. One can then notice that the 
painters vary a good deal in their artistic talents. Some paint with 
vigorous sweeping strokes, whilst others compose the picture by labo- 
riously adding one detail after another. 

If one is fortunate enough to be taken to one of these sites by an aborigine, 
he will not give a full logical explanation of the picture, but will impart 
fragments of a mental outlook which to him seems quite self-explanatory. 
But the older folk are (or at least were until recently) quite capable of 
setting forth their conceptions in a logical manner in the course of a long 
conversation. As a rule members of the younger generation cannot give 
any information about the rock pictures; nor are they any longer interest- 
ed in them, but go about their normal business, saddling and unsaddling 
their donkeys and so on, in the shade of these overhanging rocks without 
giving the pictures a single glance. 

The aborigines usually begin explanations of the rock pictures with the 
myth about the origin of the world. Members of the Unambal tribe in 
north-western Australia, for example, tell the following myth about prime- 
val times: "Heaven and earth existed from the very beginnings. They 
have 'always been there'. In the bowels of the earth there lived, and still 



Mythology 



207 



From a geographical point of view the rock pictures of Australia can be divided into two groups. 
One group comprises the rock pictures of the north-western and northern coastal regions. These 
are naturalistic in style and derive from external influences that cannot be identified with any 
accuracy. To these may be added a stylistic group in the east and south-eastern part of Australia, 
which probably also traces its origin to external influences. In the case of the second major group, 
extending across the continent from the south-west to the south east, the rock pictures are linear, 
sometimes even geometric, in style. In the mid-west and in several regions of the south-east both 
styles occur side by side or fused together. 



Ocnpelli 




208 



lives, Ungud, in the form of a large snake. Ungud is the original creative 
force from which all life develops and to which it returns after death 
to be re-born." Ungud is often also identified with the earth. Everything 
that we see and regard as the earth is really only the back of Ungud; thus 
trees grow out of Ungud's back. Ungud is also identified with water, and 
often also regarded as the concept of time and as a state of mind, 'Lalai'. 
In the sky lives Walanganda, Lord of Heaven and at the same time the 
personification of the Milky Way. Of Walanganda it is said that "he has 
made everything". At first there was nothing on earth, and only Ungud who 
lived inside it. Walanganda poured fresh water upon earth, but Ungud 
"made the water deep." Ungud also made the rain, and so life on earth 
could begin. Walanganda and Ungud never carried out their work of 
creation by day, but always by night. When they did this they were in a 
dream-like state, 'lalai'. Whilst in this state Ungud was transformed into 
the being which he, or she (for Ungud can be either male or female at 
will, and also both at once) had created or was about to create. 
Walanganda 'dreams up' the beings he creates in the same way. He is 
continually throwing down to earth, in a dream, 'soul force'. He fashions 
this 'soul force' into the image of whatever he wishes to create. He puts 
these pictures, painted in white, red and finally in black, on certain rock 
faces all over the world. This was the origin of the rock paintings to 
be found throughout the country, which are abodes for the souls of the 
beings they represent. The rock pictures are as 'fathers and brothers' to 
these beings. Only after the 'soul force' of these beings had taken shape 
in these pictures did Walanganda make the creatures themselves. Then 
he sent them out into all parts of the earth. 

In those primeval days all the beings whose image Walanganda painted 
had neither mouth nor eyes. These they were still to receive from Ungud. 
Walanganda is still going on creating and dreaming, and never lets his 
creatures on earth die out. He is constantly sending fresh 'soul force' down 
from heaven. 

In the rock pictures of the abodes of souls these figures of animals and 
plants form groups round the wondjinas. A wondjina is an an- 
thropomorphic being and the personification of rain. Ungud 'finds' the 
first wondjina in a creative dream at the bottom of the water. And just 
as Ungud found the soul of a wondjina at the bottom of the water — the 
water which is at the same time himself — so also nowadays, the aborigines 
believe, every man finds in a vision the soul of his child in the water. 
The water-holes of the country are associated with certain wondjinas. 
Every watercourse, pond or spring in north-western Australia has its 



209 




Snake with eagle. North-western Australia, 3 ft. i in. X 2 ft. ji/4 in. 



own wondjina figure situated nearby. Sometimes, however, it is a consid- 
erable distance away. A water-hole in a valley may have its wondjina 
figure painted on a rock face high up on the mountain side. 
After they had been created in the early days of the world the wondjinas 
went forth across the land. They caused the rain to fall, fashioned the 
earth, drew the course of the rivers, pushed up the mountains and flattened 
out the plains. 



At a time "when the stones were still soft" they built themselves "stone 
houses". When they 'died' they lay down upon the soft rock, leaving 
an impression behind. These 'impressions' are the rock pictures that exist 
today. To the aborigines it is of no consequence that these rock pictures 
are not impressions at all but paintings, and that they are located on sheer 
rock walls on which one cannot lie down. Of other wondjinas it is said 
that they stood in front of the rock faces and that one of their companions 
drew their silhouette, which was left behind as a shadow on the rock. At 
the spot where they left their 'impressions' behind the wondjinas dis- 
appeared into the earth, and ever since have lived at the bottom of the 
water belonging to the picture. 2 

The chief of the tribe which owes its origin to a certain wondjina and 
belongs to a certain rock picture is responsible for re-painting it before 
the rains come, to give fresh force to the soul substance contained in it 
and thus to assist in the propagation of men, animals and plants. In 
explaining how the pictures were painted, the aborigines set out their 
view of the world: talking to an aborigine in front of one of these pictures, 
one will hear him state over and over again that they only re-paint them, or 
'touch' them, but did not paint them themselves in the first place — a 
statement that seems quite natural in the light of what has been said 
above. When they 'touch' figures of wondjinas and of Ungud, they do in 
fact seem merely to touch up paintings that already exist. 
But they often paint afresh the animals, plants, fruit, celestial phenomena, 
etc., which belong to these pictures. According to information given by 
some aborigines in 1938 the fine representations of eagle-hawks at 
Kandjanlgari were only painted in very recent times. They say the same 
about a whole number of other paintings which strike one by the freshness 
of their colours. Thus it seems that the aborigines adopt a different 
attitude to the Ungud and wondjina figures on one hand and to the 



¥*&• 55 ~ Women running. 
Painting in red. Oenpelli, 
Unbalania, Northern Austra- 
lia. After Ch. P. Mountford 




211 



accompanying pictures of natural phenomena on the other. The former 
they regard as the 'shadows' of primeval hero figures, their direct legacy, 
but not the latter, apparently, for in this case man has the ability to create 
them over and over again in the course of fertility rites. Processes of 
thought which we can only follow with difficulty may play some part in 
this. Undoubtedly the aborigine is imbued with the conception of a deep 
community between himself and life in nature and the cosmos. We have 
impressive evidence of this in his primeval traditions, his belief in spirits 
of fertility and procreation, and his totemistic outlook. 3 
The significance of the rock picture galleries of north-western Australia 
in religious cults can by and large be compared with that of the tjurungas 
of Central Australia, described in detail by Strehlow, Spencer and Gillen, 
as well as with related objects among other Australian tribes. Like these 
sacred objects of wood or stone, the rock paintings may be regarded as 
the legacy of primeval times. In both cases they transmit the life-giving 
forces of great mythical figures and play an important part in fertility 
rites. The rock galleries contain paintings of primeval tribal or clan 
heroes and of the local totem. Much the same is expressed in the engraved 
motifs on the tjurungas. s * 

The wondjina figures almost always appear in combination with zoom- 
orphic and anthropomorphic representations, of which the latter seem 
to belong to different styles; in some cases they are definitely more recent 
than the wondjina figures, but in other cases they are older. Blending 



Fig. 56 — Warriors running, with spears and spcar-throwers. Oen- 
pelli, Inagurdurwil, Northern Australia. After Ch. P. Mountford 





Fig. 57 — Warriors standing, with European axes. Oenpelli, 
Unbalania, Northern Australia. After Ch. P. Mountford 



with other motifs is particularly frequent on the border of the area 
containing wondjina figures. For example, John Morgan 4 found a 
female squatting figure with the head of a wondjina by the middle 
reaches of Carson River, which flows to the west of Drysdale River: this 
is apparently a blending of two heterogeneous styles. The zoomorphic 
representations are only connected with the wondjina pictures in mythol- 
ogy. From a stylistic point of view they are different, for they are less 
rigid than the wondjina figures. In the re-touching of animal figures, too, 
greater licence is permitted, and the aborigines do not keep so closely 
to the older models still visible on the rock. 

The impulse that led to the appearance of the wondjina figure may have 
come from some area outside the Australian continent where anthropom- 
orphic representations were known. 



Plate p. 210 



213 



Plate p. 215 There is perhaps a sign of such influence in the small elegant anthropom- 
orphic figures which occur in north-western Australia as well as the 
wondjina figures. These other representations are mostly monochrome 
(from dark reddish brown to reddish yellow), and are frequently elegant 
animated figures, or groups of figures, measuring between 12 and 27 in. 
in height. Bradshaw 5 was the first to discover figures in this style on the 
Upper Regent River and to publish sketches of them. Grey, 6 the discov- 
erer of the wondjina figures, makes no mention whatsoever of this second 
elegant style — which might be termed the 'Bradshaw' or 'elegant' style. 
The reason for this omission is not that there are no such paintings in 
the Glenelg River area, but that they are so indistinct and faded as to 
be scarcely visible. Only by very thorough examination can one find traces 
on the rock walls of such representations, which were overlooked by 
Grey. Davidson 7 refers briefly to these pictures but refrains from passing 
any judgment on them, considering that the information in his possession 
was inadequate. 

An attempt to classify stylistically the numerous anthropomorphic figures 
in north-western Australia which do not belong to the wondjina style 
would give a result roughly as follows: 

Plate p. 215 The first and most striking group comprises the few very elegant small 
figures in profile or semi-profile; the very graceful lines convey a sense 
of movement; and the mode of representation seems so 'un-Australian' 
as to suggest that they are associated with some other culture. 
Related to this group are the figures with very exaggerated hairdress. 
They often appear to be clothed, have 'wide sleeves', and carry bags in 
their hands. This group also includes the figures reproduced by Bradshaw. 
At the rock picture sites in Eastern Kimberley in particular, but also in 
Western Kimberley, there occur other figures which can be assigned 
neither to the wondjina style nor to any of the groups of the so-called 
'Bradshaw style'. They are rather grotesque figures, sometimes male but 
generally female, which are still painted nowadays. These 'demon 
figures' are the last remnant of a tradition which one comes across quite 
• frequently. These are the so-called 'squatting figures' (mostly female, as 
already mentioned), which seem to degenerate in the course of time from 
rather strict stylization to grotesqueness. In addition to reproductions 



Small figures with arcs over their heads. An illustration of the 'elegant style'. Wonalirri, North- ► 
western Australia. / ft. ii/4 in. x 7 in. 

214 




215 



of this kind, obviously quite recent, there are also a large number of 
older figures — e.g., on the roof of one site containing rock pictures, and 
even older ones at the foot of one wondjina figure. 
Plate p. 219 At Ngungunda 2 a strictly stylized squatting figure is portrayed, with 
beams radiating from the head and the body filled in with parallel 
stripes; aborigines who saw it eventually designated it a 'she-devil', 
"because she-devils are always in the habit of lying down in this indecent 
manner" — the indecency being indicated by contemptuous laughter on 
the aborigines' part. 8 They said that the squatting figure actually rep- 
resented the act of coitus, a woman lying on her back with her legs wide 
apart. In Western Kimberley Mrs. Schulz 9 came across similar pictures 
and received the same sort of information from the aborigines. 
A fourth group comprises all those figures that defy classification because 
they are unique. 

Common to all groups, apart from the fact that they differ distinctly from 
those in the wondjina style, is the almost complete absence of zoomorphic 
representations. Bradshaw was the only investigator to find his figures 
together with those of animals — but judging from the sketches he 
published the animals may be later addenda. 

Worms 10 holds the view that there must once have been far more pictures 
than there are now, since only those in protected places could be preserved, 
and innumerable others must have been effaced by weathering — by the 
heavy monsoon storms and the glowing sunshine. Worms compares these 
representations with the wondjina pictures and comes to the conclusion 
that the latter served a mythological purpose whereas the giro giro pictures 
(a native term for the graceful paintings with the small figures) were 
representations of the daily life of an older race now extinct. He has in 
mind a pre-australoid, perhaps negrito-tasmanoid race. 
If, therefore, one attempts to establish the connection between the two 
styles in north-western Australia from the standpoint of historical evolu- 
tion, one finds that an important process took place — despite the fact 
that the anthropomorphic wondjina figures became more primitive in 
the course of time as compared with the 'elegant (Bradshaw) style' paint- 
ings. The latter do not give the impression of having been the focus of 
a cult, but appear to have been produced to satisfy pleasure in art for 
its own sake. The wondjinas, on the other hand, with their naive nat- 
uralism, are often painted in beautiful colours and are well adapted to 
the uneven surface of the rock; the whole mental outlook of the aborigines 
revolves around them. Here an antropomorphic style has been introduced 
from outside Australia and incorporated into an ancient conceptual 

216 



F'g- 58 ~ Anthropomorphous 
figures copulating. Pecked out 
on rock. Pilgangura, Abydos, 
North-west Australia 





I 1 filt 
iy i Mi. ȣ& K 



? a s 6 *¥•>. „' 



\ 



\ <8Sfe._/ 



framework, in which this style of painting was originally unknown. The 
creative force Ungud, still today thought of as a snake, was originally 
also represented as one; and it was only under the influence of some 
external model that it was also conceived of and portrayed in human form. 
Although the anthropomorphic style became much more primitive, as 
a result of this process of assimilation, this process shows that the 
aborigines of north-western Australia manifested very great aptitude in 
absorbing these influences from abroad and maintaining the new style 
for a long time afterwards. The development of the wondjina figures 
out of the 'elegant style' figures must have taken place a fairly long time 
ago. This is clear from the mere fact that a comparison between the 
present wondjina figures and those discovered by Grey in 1838 shows that 
no stylistic changes took place during the intervening century. It is 
unlikely that, once the 'elegant style' had developed into the wondjina 
style, one element of the former style should have continued to exist. 
The numerous animated human figures in a degenerate style that are 
still being painted nowadays, and which can be found around (and 
sometimes in) the wondjina pictures, cannot be derived from the 'elegant 
style'. They were more probably stimulated from the east, by the in- 
fluences radiating from the second great centre of rock art on the northern 
periphery of the Australian continent, Oenpelli, in Arnhem Land. 



217 



Rock pictures To the east of the Kimberleys, by Forest River, there are still traces of 
rn em an the won djina style. Basedow, 11 who studied these rock pictures, published 
a reproduction of a wondjina head and animal paintings which invite 
comparison with those of the Kimberleys, as well as anthropomorphic 
figures which can only be equated with the last degenerate paintings of 
the north-west. 

Towards the east the wondjina style no longer appears in its pure form, 
and in the Northern Territory we only find sporadic sites where the style 
of the pictures can be related to that of the paintings of the Kimberleys. 
We are again indebted to Basedow 12 for some re-touched rock pictures 
from the area of Humbert River, a tributary of Victoria River, and from 
Pigeon Hole. They are represented in a curious vigorous style and are 
probably not related to the Kimberleys paintings. The figure of a squat- 
ting or running woman from Blunder Bay can at best be regarded as 
equivalent to the demon figures from the Kimberleys. 
Distinct affinities with the Kimberleys paintings are, however, again found 
at Delamere, the site discovered and studied by Davidson. The affinities 
with the wondjina pictures of the west are not only external, in the 
arrangement and placing of the pictures, but, as Davidson has established, 
also extend to the myths they represented. The anthropomorphic figures are 
known as 'lightning brothers'. The resemblance to the wondjinas also 
extends to characteristics such as the filling in of the body with stripes, 
the absence of a mouth, and the beams or rays framing the head. This 
site contains very primitive animal figures and both from a geographical 
as well as from a stylistic point of view seems to lie between the two art 
centres of the north, the Kimberley area and Arnhem Land. 
Apart from Delamere and probably some eastern wondjina sites as well, 
the influence of the great centre of rock art at Oenpelli makes itself felt 
in the Victoria River and Humbert River area, and still more in the 
Katherine River area further to the east. In 1947-8 C. P. Mountford, 13 
leader of a joint Australian-American expedition, discovered near this 
missionary station several large rock art sites. They contain representa- 
Fig. 54 tions of animals, especially fish, turtles and snakes, most of which have 
a marked 'X-ray style', i.e. internal organs not normally visible, such as 
Figs. 55, 56 the stomach, heart, lungs or spine, are painted as well. With these pictures 
the aborigines still have an active relationship, but this is less the case 
with the innumerable anthropomorphic figures covering the rock walls, 
which they believe to be the work of spirits. The style of these figures is 
curiously diversified — at once vivid and degenerate. The representations 
comprise human beings in profile, running carrying a spear and spear- 

218 




Squatting figure. Ngungunda 2, North-western Australia. 2 ft. 1/2 !n - X 1 ft. 7 in. 



219 



thrower, and also groups of men standing or dancing, often with spindly 
limbs. In most cases several layers and styles are superimposed upon one 
another. It is, however, scarcely possible to establish a chronological se- 
quence at Oenpelli; at one site certain types can be identified as more 
recent, but at others the evidence points the other way and upsets one's 
tentative conclusions. Nor can the style in which animals are represented 
automatically be regarded as more recent, because there are some rep- 
resentations of animals on which human figures are superimposed and 
which must therefore be older. One approach to an understanding of 
this style may perhaps be by way of the bark paintings still executed today 
in Arnhem Land. 
influences radiating Apart from the rock pictures closely related in style to those of Oenpelli 
from Oenpelh t h ere are others to the south, south-west and probably also east of this 
centre where the anthropomorphic influences radiating from it are less 
marked, and finally are no longer to be found at all. The border of the 
zone influenced by Oenpelli seems to lie along the Roper River in the 
south and the Katherine River in the west. Further east, on Groote 
Eylandt, Tindale 14 discovered further rock pictures, composed of an- 
thropomorphic and zoomorphic motifs. They can be compared with the 
bark paintings still produced nowadays on the island and do not appear 
to be very old. They could, however, be regarded as a last eastern offshoot 
of the Oenpelli anthropomorphic representations, but the reasons for 

F'g- 59 ~ Figures resembling human beings, peck- *,. r $i- _ A ,"> 

ed out on roch. Depuch I., Western Australia i£-\. V "V- "**.■ VS St JiA 










Fig. 60 — Hunting scene. Rock engraving. Gos- 
ford near Sydney. N.S.W. After D. S. Davidson 



^EQ^-4- 



supposing that such a link existed are based on geograph- 
ical considerations rather than stylistic affinities. All the 
rock pictures are merely primitive full-face figures such 
as are also found in south-eastern Australia. McCarthy 1 5 
found animal pictures on Groote Eylandt. These are 
also comparable stylistically to the bark paintings in 
which hunting and fishing scenes are depicted, but are 
more advanced. 

A fruitful comparison may be drawn between the rock 
art of the Oenpelli area and that of the area in the south 
subject to its influence, as well as that of the north-west; 
a detailed comparison between the 'elegant style' of the 
Kimberleys and some pictures from Oenpelli shows that 
they have common features. But at the same time some 
vital differences must be pointed out, which show that 
although the two areas may have responded to identical 
stimuli from without they followed completely different 
lines of development. South of King Leopold Ranges 
pictorial art as found in the north-west and the Kimber- 
ley area suddenly comes to a stop. Only in the northern 
part of Western Australia, particularly in the Yule River 
area, do we still find engravings or peckings of an- 
thropomorphic and zoomorphic figures. In the south aboriginal art is 
represented only at isolated sites by increasingly abstract linear figures, 
often very primitive in style. Both styles, the abstract and the realistic, 
occasionally overlap. South-east of Arnhem Land, in the north of Queens- 
land, we only come across occasional engravings or paintings of impor- 
tance, and their quality can no longer be compared with those of northern 
and western Arnhem Land. 

In New South Wales the best-known works are the rock engravings, some 
of them of gigantic size, not far from Sydney. These simple outline draw- 
ings are very plentiful, often measure as much as 20 to 30 feet in length, 
and are for the most part situated on broad ledges at a high altitude. 
They are among the first rock pictures to have been studied, and there is 




New South Wales 




Fie. 60 



Fig. 61 — Painting in red and yellow. 
Walgarock, Western Australia. After 
D. S. Davidson 



already a considerable literature 
on the subject. McCarthy 10 has 
recently been engaged in compil- 
ing an exact catalogue of these 
rock pictures, the number of 
which is constantly growing as 
new finds are made. For the most 
part they represent living crea- 
tures (fish, kangaroos, ostriches 
and birds), utensils and weapons (boomerangs and shields), and numerous 
anthropomorphic representations, mostly male — all executed in a prim- 
itive yet spirited manner. Sometimes the bodies are filled in with vertical 
lines; the faces are generally suggested by eyes and the hair by disconnect- 
ed radiating lines; mouth and nose are absent. One particular posture — 
the legs wide apart and 'squatted off' — is obviously intended to express 
movement, probably in dancing. There can scarcely be any doubt that the 
rock engravings near Sydney had a ritual significance and that the large 
compositions to be found here represented mythological concepts. But it 
has not been possible to obtain information about the rock pictures from 
the aboriginal population, since in this area they died out long ago. In 
addition to these engravings paintings are also found in New South 
Wales, but these are scarce and very primitive in their manner of 
execution. Davidson 7 gives some examples of those found in Milton 
District. In the eastern part of New South Wales extraordinarily large 
and rather ungainly paintings have been found, which in all probability 
represent deified culture heroes and progenitors of tribes that are now 
extinct. 

In the western part of New South Wales there is a wealth of rock engrav- 
ings and paintings. An account of these sites has been published by 
Lindsay Black. 17 Mc Carthy is covering this region as well in his detailed 
inventory. 

In Victoria, according to Davidson, paintings are very rare and no 
engravings at all have as yet been discovered. But the paintings that exist 
are nevertheless noteworthy. In the first place there is the site of Glen Isla, 



222 



where the pictures were copied by Matthew in 1897 and are reproduced 
by Davidson. 7 But Matthew's copies are apparently not faultless. This 
opinion is endorsed by Adam, 18 who has studied this site in detail; but 
it would be difficult to improve on them now, since the colours have 
obviously faded very much since 1897. 

The other site of rock pictures in Victoria is at Mount Langi Ghiran. 7 
It represents a figure holding a "disproportionately large" boomerang. 
Davidson sees no reason why this site should be called 'the cave of the 
snake' by the aborigines, but Mrs. Schulz, who visited this site in 1956, 
provisionally identified the 'boomerang' as a snake, which would explain 
this designation. 
The most important site in Victoria is the Conic Range rock-shelter, 











?3* 1 * 








■ ! M> '- • 
















■ 




W?m 






II 






EMG^M/IKx 




wSmk 








f 


iM^fx^M 


\ <' ' ■ 


MHa '£tm 


'^Bm fjL i 


/ 










1 




















K^BTJt^HNH 




L 


F^thH 


i\ A 





Lizard-like animal. Rock painting. Hawker, Southern Australia. 3 ft. 2 in. 



223 



an account of which has been published by Tugby. 19 These figures of 
Fig. 64 slender human beings, painted in dark red, bear a strong resemblance 
to those in the 'elegant style' in north-western Australia. But there, too, 
the picture seems to have been badly damaged by weathering, and it is 
difficult to make out the details, ft must be recognized that the three 
sites of rock art in Victoria are unusual from a stylistic point of view, and 
cannot simply be grouped together with the rock pictures of nearby New 
South Wales or of Southern Australia. 
Southern and In Southern Australia — at Devon Downs, for example — there are 
Central Australia labyrinth-like engravings. Type B from Devon Downs has been subjected 
Fig. 6- to radio-carbon analysis and gave the date of approximately 2200 B.C. 20 
A more recent development at the same site has finer simpler lines 
and, perhaps, also human figures reduced to symbols, such as are fre- 
quently found in rock pictures in Southern and Central Australia. 
Labyrinth-like forms also occur as engravings or peckings at Burra and 
in the Flinders Ranges in Southern Australia, and extend to the centre 
of the continent. 
Figs. 62, 63 In the Flinders Ranges peckings of animal or (apparently) human 
figures are also met with. They might perhaps be described as a reminder 
of the figure at Mootwingee in the west of New South Wales. But in 
addition we find here concentric circles with cross-lines such as are also to 
be seen at Devon Downs. Perhaps these suggest that the influence of 
Central Australia was felt here. The large sites at Eucolo Creek near 
Pimba contain concentric circles, the comb motif, plant motifs and what 
are clearly anthropomorphic figures. 21 

Anthropomorphic representations in cruder form are to be found at 
South Para River and in a simplified abstract form at Malkaia. 7 It is 
not clear whether the human figures in the paintings of Southern 
Australia should be traced back to the influence of anthropomorphic art 
from Victoria or New South Wales, or whether they denote influences 
from the north, which might have reached Southern Australia via the 
centre, bypassing the west, where there are no anthropomorphic repres- 
Plate p. 227 entations. If that is so, a connecting link would be provided by the head 
with beams radiating from it at Ayers Rock. The rather clumsy figures 
from South Para River are reminiscent of those at Delamere as well as 
the animal figures at Ayers Rock. It seems clear enough that influences 
from the north reached the centre and the south by way of the Western 
Australian desert. The mythological evidence goes to show that in 
Western Australia the aborigines still recollect a cultural movement from 
the west across the desert to the coast of the Great Australian Bight. 22 

224 



Fig. 62 — Human and animal 
figures. Euriouve. western N.S.W. 




The fact that such striking implements as the threadcross were used both 
by the tribes of Western Australia, as well as the linguistic affinities that 
exist between them, suggest that they are related to one another. 
It would also be possible to link the labyrinth-like representations in 
Western with those in Southern Australia, thus making for a homog- 
eneous area in the south-west of the continent where this motif appears. 
It can be assumed that influences radiated from the centre to the north- 
west and north. In this region the representation of a snake in the form 
of a spiral seems to be derived from the spiral style of the centre. Paintings 
of this kind are to be found at Yule River in Western Australia, in the 
McDonnel Ranges, in Central Australia, at Ooraminna, at Delamere, 
on a painted sheet of bark from Liverpool River and at Ngungunda in 
the Kimberleys. 8 

In Central Australia, and in a similar form in Western Australia as well, 
there is a style of rock art that involves abstract lines. The meaning of 
these drawings cannot be established. It is clear that in Central Australia, 
as well as in the centre of Western Australia under the influence of the 
centre, concentric circles often serve as the starting point of such lines, 
but already in the western part of Central Australia such spirals and 
circles are driven into the background: here a drawing may often have 
no more than one spiral or circle. In the southern part of Western 






225 



Australia no spirals are found at all, but only labyrinth-like lines, arranged 
in an apparently arbitrary fashion. 
Conclusions This general survey of Australian art leads to certain interesting results. 
In the first place we see that the group of rock pictures consisting of 
naturalistic anthropomorphic and zoomorphic figures can be contrasted 
with a second group that gives preference to geometric signs, which is 
to be found in Western, Central and Southern Australia. There are, of 
course, points of transition between these two groups, and for this reason 
no hard-and-fast geographical dividing-line can be drawn between them. 
But the geometric style of the rock pictures of the south and west seems 
to be concentric, and is perhaps older. It must be borne in mind that at 
the present day the style of the human figures on the bark paintings in 
the north is steadily being superseded — smothered, so to speak — by a 
geometric style, the main motif of which is a rhomb, but which also 
includes the herring-bone pattern. In the ceremonial style of the bark- 
paintings in Arnhem Land, which differs from the 'ordinary' style 
chiefly by the use of geometric motifs, a preference seems to be shown 
for geometric symbols, as is also the case with the tjurungas in Central 
Australia. These motifs to be found on the northern periphery of the 
continent are thought to be older, since they are ceremonial — particularly 
the so-called herring-bone and the lozenge patterns. It is quite clear that 
these motifs are not merely supplementing anthropomorphic and 
zoomorphic ones, but are gradually crowding them out. The old ceremo- 
nial motifs are gaining the ascendancy and the naturalistic-realistic style 
is giving way to one that is geometric and abstract. 




Fig. 63 — Human and animal 
figures. Euriowie, western N.S.W. 



226 







\*t*% 



, A 



Rock painting. Avers Rock, Central Australia. Width -j ft. 11 1/4 in. 



The naturalistic style of rock art extends in an arc from the furthermost 
north-west to the east and south-east of the continent. On both sides of 
this stylistic 'bridge', which is clearly weaker in the middle than at either 
end, there are paintings and engravings in which the two styles are merged 
— as, for example, in the few rock paintings of Queensland and the eastern 
part of New South Wales, and in the engravings in the eastern part of 
Southern Australia. 

The naturalistic rock paintings can be divided into different sub-groups 
according to their motifs and their quality as works of art. In the Kim- 
berleys two styles can clearly be distinguished: the wondjina style and 
the so-called 'elegant style'; it seems possible that the former, which still 
survives today, is derived from the latter. Traces of the wondjina style 
are met with far away in Southern, Western and Central Australia. The 



227 



Fie. 54 



style of the second artistic centre in the north, Oenpelli, is comparable 
in some phases to the 'elegant style'. There is no reason to dispute the 
view that these rock pictures were the focal point of cults, as we may 
do in the case of the Kimberleys paintings. The sites in Victoria and in 
the western part of New South Wales resemble in style those of Oenpelli. 
But there are no intermediate stations which could lend support to the 
theory that this style spread over the whole of Australia. 
In both areas, the Kimberleys and Oenpelli, zoomorphic motifs appear 
together with anthropomorphic ones, but it seems that they only do 
so at a stage when anthropomorphic representation begins to flag. In 
the Kimberleys they seem to be connected with the wondjinas; but the 
mythological evidence suggests that this connection has not always 
existed and that in recent times anthropomorphic figures have been 
superseded by zoomorphic ones, particularly snakes. 
It is also clear that these two areas occupied a special position as centres 
from which foreign influences were transmitted to the rest of the con- 
tinent. Anthropomorphic representation spreads outwards from the 
Kimberleys and Oenpelli. Perhaps in both cases the original stimulus 
may have come from without. But Oenpelli is. unique as regards the 
representation of animals in 'X-ray style'. In the south-east a similar 
sub-division can be made, although the rock pictures never attain the 
same level of artistic accomplishment as do those of the north-west. But 
here, too, in Victoria and the western part of New South Wales, we 
occasionally find small anthropomorphic paintings. The large anthropom- 
orphic figures (paintings in Victoria and at isolated sites in the north- 
eastern part of New South Wales, and engravings in the vicinity of Sydney) 
are found in combination with primitive animal figures and are clearly 
degenerate forms of a more advanced art, probably the older style with 
the small figures. 
These conclusions are, moreover, substantiated by study of the quality 




Fig. 64 — Squatting figure. Paint- 
ing in red, Conic Range, eastern 
Victoria. After D. J. Tugby 



Fig. 65 — Rock engraving. Devon Downs 
Southern Australia. After D. S. Davidson 




of the works. As a rule it may be said that nowadays Australia appears 

to be in the grip of a tendency towards schematization, which is cramping 

all artistic expression, so that works of art charged with living tension 

are very rare. A map could be drawn showing the location of these few 

inspired works. One would then see that in the sphere of geometric or 

abstract art such motifs only occur — or occurred until about a century 

ago — in the extreme west and south-west on ceremonial staffs. Perfect 

specimens of these could still be collected by Europeans. In the centre, 

north-east and north they are extremely rare. But in the centre a few rock Fig. 65 

paintings have been preserved in which the geometric or abstract style 

is shown to perfection. The only rock engraving of ascertainable date — 

that at Devon Downs — which was produced in the third millennium 

B.C., is markedly inferior. 

In anthropomorphic art this tension was only able to survive in the 

oldest examples and in the north-west, the area where contact is thought 

to have been made with the outside world. Anthropomorphic art continues 

up to the present day, but all of it, both rock art and bark paintings, seems 

degenerate, lacking in inspiration and vigour of artistic expression. 




Fig. 66 — Rock painting. Malkaia, Southern 
Australia. Af'cr D. S. Davidson 



229 



1 Elkin, A.P., Grey's Northern Kimberley cave-paintings re-found, in: Oceania, 
Vol. XIX, No. i, pp. 1—15, Sydney, 1948. 

2 Lommel, Andreas, Die Unambal, ein Stamm in Nordwest-Australien, in: Mono- 
graphien zur Volkerkunde, herausgegeben vom Hamburgischen Museum fur 
Volkerkunde, No. II, Hamburg, 1953. 

3 Petri, Helmut, Sterbende Welt in Nordwest-Australien, in: Kulturgeschichtliche 
Forschungen, Vol. 5, Brunswick, 1954. 

3 a Ibid., pp. 129—216. 

4 Morgan, John, Unpublished sketches for a new map of North-western Australia. 

5 Bradshaw, Jos., Notes on a recent trip to Prince Regent River, in: Proceedings 
of the Royal Geographical Society of Australasia, Victoria Branch, Vol. 9, pp. 99— 
100, Melbourne, 1892. 

6 Grey, George, Journals of two expeditions of discovery in North-west and 
Western Australia during the years 183J, 1838 and 1839, London, 1841. 

7 Davidson, D. Sutherland, Aboriginal Australian and Tasmanian rock carvings 
and paintings, in: Memoirs of the American Philosophical Society, Vol. V, 
Philadelphia, 1936. 

8 Lommel, Andreas and Katharina: Die Kunst des fiinften Erdteils, Museum fur 
Volkerkunde, Munich, 1959. 

9 Schulz, Agnes Susanne, North-west Australian rock paintings, in: Memoirs of the 
National Museum of Victoria, No. 20, pp. 7—57, Melbourne, 1956. 

10 Worms, Ernest A., Contemporary and prehistoric rock paintings in Central and 
Northern North Kimberley, in: Anthropos, Vol. 50, Nos. 4—6, Fribourg, 1955. 

11 Basedow, H., The Australian aboriginal, Adelaide, 1925. 

12 Basedow, H., Anthropological notes on the West Coastal tribes of the Northern 
Territory of Australia, in: Transactions of the Royal Society of South Australia, 
Vol. 31, 1907. 

13 Mountford, Charles P., Art, myth and symbolism, in: Records of the American- 
Australian Scientific Expedition to Arnhem-Land, 1948, Vol. I, Melbourne, 1956. 

14 Tindale, Norman B., Natives of Groote Eylandt and of the West Coast of the 
Gulf of Carpentaria, in: Records of the South Australian Museum, Vol. II, No. 2, 
1946. 

15 McCarthy, Frederick D., The cave paintings of Groote and Chasm Islands, in: 
Records of the American-Australian Scientific Expedition to Arnhem-Land, 1948, 
Vol. II, Melbourne, 1959. 

16 McCarthy, Frederick D., Various articles. 

17 Black, Lindsay, Aboriginal art galleries of Western New South Wales, Pt. Ill, 
Melbourne, 1943. 

18 Adam, Leonhard, Rock paintings near Glen Isla, Victoria, in: Mankind, Vol. 
IV, No. 8, pp. 343-345. Sydney, 1952. 



19 Tugby, Donald J., Conic Range Rock Shelter: a preliminary account of rock 
paintings in. North-east Victoria, in: Mankind, Vol. IV, No. 11, pp. 446—450, 
Sydney, 1953. 

20 Tindale, Norman B., Culture succession in South-east Australia from Late 
Pleistocene to the present, in Records of the South Australian Museum, Vol. XIII, 
No. 4, pp. 1—49, Adelaide, 1957. 

21 Hall, F.J.; McGowan, R.G.; Guleksen, G.F.; Aboriginal rock carvings: a locality 
near Pimba, S.A., in: Records of the South Australian Museum, Vol. IX, No. 4, 
Adelaide, 1951. 

22 Basedow, H., Anthropological notes made on the South Australian Government 
North-west Prospective Expedition, 1903, in: Transactions of the Royal Society of 
South Australia, Vol. 28, 1904. 



231 



APPENDICES 



BIBLIOGRAPHY 



FRANCO - CANTABRIAN ROCK ART 

Alcalde del Rio, H., Breuil, H. and Sierra, L., Les 
cavernes de la region cantabrique, 2 vols., Mo- 
naco, 1911. 

Begouen, Cte. H., La grotte de Baume-Latrone, 
in: Mem. de la Soc. Arch, du Midi de la 
France, XX, 194-1, pp. 101—130. 

Breuil, H., Les peintures de la caverne de Niaux 
(Ariege), in: Bull, de la Soc. prehist. de l'Ariege, 
1950. 

Breuil, H., La caverne ornee de Rouffignac (Dor- 
dogne), in: Memoires de l'Academie des In- 
scriptions et Belles-Lettres, vol. XLIV. 

Breuil, H., Obermaier, H. and Alcalde del Rio, 
H., La Pasiega a Puente Viesgo (Santander), 
Monaco, 1913. 

Breuil, H. and Obermaier, H., The cave of Alta- 
mira at Santillana del Mar, Spain, Madrid, 

1935- 

Breuil, H., Obermaier, H. and Verner, W., La 
Pileta, Monaco, 1915. 

Capitan, L., Breuil, H. and Peyrony, D., Les figu- 
res gravees a l'epoque paleolithique sur les 
parois de la grotte de Bernifal (Dordogne), in: 
Revue de l'Ecole d'Anthrop., 1903, pp. 202— 
209. 

Capitan, L., Breuil, H. and Ampoulange, Une 
nouvelle grotte paleolithique a parois gravees: 
la grotte de la Greze, in: Revue de l'Ecole 
d'Anthrop., 1904, pp. 487—494. 

Capitan, L., Breuil, H. and Peyrony, D., La ca- 
verne de Font-de-Gaume aux Eyzies (Dordog- 
ne), Monaco, 1910. 

Capitan, L., Breuil, H. and Peyrony, D., Les 
Combarelles aux Eyzies (Dordogne), Monaco, 

19=4- 

Capitan, L., Breuil H., Peyrony, D. and Bourrinet, 
Les gravures sur cascade stalagmatique de la 
grotte de la Mairie a Teyjat, in: XlVe Con- 
gres Intern. d'Anthrop. et d'Arch. Prehist., 
Geneva, 1912. 

Cartailhac, E. and Breuil, H., Les peintures et 
gravures murales des cavernes pyreneennes, IV: 
Gargas, Aventignan (Hautes-Pyrenees), in: 
L'Anthropologie, 1910, p. 129. 

Daleau,,F., La grotte de Pair-non-Pair, Associa- 
tion Francaise pour l'Avancement des Sciences, 
Montauban, igo2, pp. 786—789. 

Craziosi, P., Les gravures de la grotte Romanelli, 
in: IPEK, VIII, 1932-3. 



Graziosi, P., Le Pitture e graffiti preistorici dell' 
isola di Levanzo nell'Arcipelago delle Egadi 
(Sicilia), in: Rivista di Scienze Preistoriche, 
1950. 

Lalanne, G., Bas-Relief a figurations humaines de 
l'abri sous roche de Laussel (Dordogne), in: 
L'Anthropologie, 1912, pp. 129—148. 

Lalanne, G. and Breuil, H., L'abri sculpte du Cap 
Blanc a Laussel (Dordogne), in: L'Anthropolo- 
gie, 1911, pp. 385-402. 

Lemozi,A., La grotte-temple du Pech-Merle, Paris, 

1929- 
Martin, H., Les sculptures du Roc de Sers, in: 

Prehistoire, I, 1932, pp. 1—8. 
Passemard, E., La caverne d'Isturitz en pays 

basque, in: Prehistoire, September 1944. 
Trombe, F. and Dubuc, G., Le centre prehistorique 

de Gantier-Montespan, in: Arch, de l'Inst. de 

Paleont. Humaine, Paris, 1947. 
Vezian, J., La grotte du Portel, in: Bull, de la 

Soc. Prehist. du Languedoc, no. 2, 1945, pp. 

2—11. 
Windels, F., Lascaux: 'Chapelle Sixtine' de la pre- 
histoire, Montignac (Dordogne), 1949. 

THE ROCK PICTURES OF THE 
SPANISH LEVANT 

Almagro, M., La cronologia del arte levantino de 
Espafla, in: Congres Int. des Sciences Pr£- 
historiques et Protohistoriques, Actes de la 
IHe Session, Zurich, 1950, Zurich, 1953, pp. 
142-9. 

Almagro, M., El Covacho con Pinturas rupestres 
de Cogul (Lerida), Lerida, 1952. 

Almagro, M., Las pinturas rupestres levantinas, 
Madrid, 1954. 

Bandi, H.-G. and Maringer, J., Kunst der Eiszeit 
2nd ed., Basle, 1955. 

Breuil, H., Les roches peintes de Minateda (Alba- 
cete), in: L'Anthropologie, vol. 30, 1920, pp 
1-50. 

Breuil, H., Serrano Gomez, P. and Cabre Aguili 
J., Les Abris del Bosque a Alpera (Albacete) 
in: L'Anthropologie, vol. 23, 1912, pp. 529— 
562. 

Cabre Aguild, J., El arte rupestre en Espana, Ma- 
drid, 1915. 

Hernandez-Pacheco, E., Las pinturas prehistoricas 
de las cuevas de la Arafia (Valencia), Madrid, 
1924. 



234 



Kiihn, H., Kunst und Kultur der Vorzeit Europas, 

Vol. I: Das Palaolithikum, Berlin, 1929. 
Lindner, K., Die Jagd der Vorzeit, Berlin and 

Leipzig, rg37. 
Narr, K. J., Das hohere Jagertum: Jiingere Jagd- 

und Sammelstufe, in: Historia Mundi, Vol. I, 

Berne, 1952. 
Narr, K. J., Interpretation altsteinzeitlicher 

Kunstwerke durch volkerkundliche Parallelen, 

in: Anthropos, Vol. 50, 1955, pp. 513-545- 
Narr, K. ]., Vorderasien, Nordafrika und Europa, 

in: Abriss der Vorgeschichte, Munich, 1957. 
Obermaier, H., Nouvelles Etudes sur l'art rupestre 

du Levant Espagnol, in: L'Anthropologie, Vol. 

47. 1937. PP-477-49 8 - 
Obermaier, H. and Breuil, H., Las pinturas ru- 

pestres de los alrededores de Tormdn (Teruel), 

Madrid, 1927. 
Obermaier, H. and Wernert, P., Las pinturas 

rupestres del Barranco de Valltorta, Castellon, 

Madrid, 1919. 
Obermaier, H. and Wernert, P., La edad cuater- 

naria de los pinturas rupestres del Levante 

Espanol, in: Mem. de la Real Sociedad Espa- 

nola de Historia Natural., Vol. XV, Madrid, 

»929. PP- 527-537- 
Pericot-Garcia, L., La cueva del Parpall6 (Gandia), 

Madrid, 1942. 
Pidal, R. M. (ed.), Historia de Espafia, Vol. I, 

Madrid, 1954. 
Porcar, J. B., Obermaier, H. and Breuil, H., Las 

pinturas rupestres de la Cueva Remigia, Ca- 

stelkm, Madrid, 1936. 
Vilaseca, S., Las pinturas rupestres de la cueva del 

Polvorin, Madrid, 1947. 

ROCK ART OF THE MAGHREB 
AND SAHARA 

Almagro, M., Prehistoria del Norte de Africa y 
del Sahara espanol, Instituto de Estudios afri- 
canos, Barcelona, 1946. 

Breuil, H., in collaboration with Henri Lhote, 
Les roches peintes du Tassili-n-Ajjer, Paris, 
Arts et Metiers graphiques, 1954. 

Chasseloup-Laubat , F. de, Art rupestre au Hog- 
gar, Paris, Plon, 1938. 

Dalloni, M., Mission du Tibesti, in: Mem. Ac. 
Sciences, Pt. II, Paris, 1935. 

Flamand, G. M. B., Les Pierres ecrites, Paris, 1921. 

Frobenius, L. and Obermaier, H., Hadschra Mak- 
tuba, Munich, 1925. 

Frobenius, L., Ekade Ektab: die Felsbilder Fez- 
zans, Leipzig, 1937. 



Graziosi, P., Arte rupestre della Libia, Naples, 

1942- 

Lhote, H., Gravures, peintures et inscriptions ru- 
pestres du Kaouar, de FAir et de TAdrar des 
Iforas, in: Bull. Inst. Franc. Afrique Noire, 
Vol. XIV, no. 4, Oct. 1952, pp. 1268—1340. 

Lhote, H., Peintures rupestres de l'oued Takeche- 
rouet (Ahaggar), in: Bull. Inst. Franc. Afrique 
Noire, Vol. XV, no. 3, 1953, pp. 283—291. 

Lhote, H., Le cheval et le chameau dans les pein- 
tures et les gravures rupestres du Sahara, in: 
Bull. Inst. Franc. Afrique Noire, Pt. XV, no. 
3. J ul Y '953. PP- ii3 6 -!228. 

Lhote, H., La route antique du Sahara Central, 
in: Encyclopedic mensuelle d'Outre-Mer, Vol. 
I, Nov. 1951, p. 300. 

Lhote, H.. Die Felsbilder der Sahara, Wiirzburg- 
Vienna, 1958. 

Monod, Th., LAdrar Ahnet, Institut d'EthnoIo- 
gie, Paris, 1932. 

Rhotert, H., Libysche Felsbilder, Darmstadt, 1952. 

Vaufrey, R., L'art rupestre nord-africain, Arch. 
Inst. Paleont. Humaine, Paris, 1939. 

ROCK ART OF SOUTH AFRICA 

Most of the studies of South African prehistoric 
art are articles in learned journals; they are listed 
here without the title being given. This is done 
only in the case of book-length works. 
Armstrong, A. I., Journ. Roy. Anthrop. Inst., 

1931, Vol. 61, pp. 239-276. 
Bartels, M., Zeitschr. f. Ethnol., 1892, Vol. 24, 

pp. 26-7; Vol. 25, p. 32. 
Battiss, W. W., The Amazing Bushman, Pretoria. 
Battiss, W. W., The Artists of the Rocks, Pretoria, 

1948. 
Bleek, D. F., S. Afr. Journ. Sci., 1932, Vol. 29, pp. 

72-83. 
Bleek, W. H. I. and Bleek, D. F., The Mantis and 

his Friends, Cape Town, 1923. 
Bleek, W. H. I. and Lloyd, L. C, Specimens of 

Bushman Folklore, London, 1911. 
Bleek, W. H. I., Cape Monthly Mag., N.S., 1874, 

Vol. 9, pp. 10-13. 
Breuil, H., The White Lady of the Brandberg, 

London, 1955. 
Breuil, H., Les Roches Peintes d'Afrique Australe, 

Paris, 1954. 
Breuil, H., Philipp Cave, London, 1957. 
Breuil, H., The Tsisab Ravine, Clairvaux, 1959. 
Breuil, H., Man, 1930, Vol. 30, pp. 149—151; S. 

Afr. Arch. Bull., 1945, Vol. I, No. 1, pp. 5—7; S. 

Afr. Journ. Sci., 1945, Vol. 41, pp. 353—5; Man, 



235 



1946, Vol. 46, p. 84; L Anthropologic, 1949, 
Vol. 53, pp. 377—406; France Abroad, Vol. 3, 
No. 9; S. Afr. Arch. Bull., 1949, Vol. 4, No. 13, 
pp. 14—18; id., pp. 19—27; Vol. 4, No. 14, pp. 

39-5°; etc - 

Burkitt, M. C, South Africa's Past in Stone and 

Paint, Cambridge, 1928. 
Christol, F,, L'Art dans l'Afrique Australe, Paris, 

1911. 
Dart, R. A., Nature, 1925, Vol. 115, pp. 425—9; S. 

Afr. Journ. Sci., 1931, Vol. 28, pp. 475—486. 
Dornan, S. S., Pygmies and Bushmen of the Kala- 
hari, London, 1925. 
Dornan, S. S., Journ. Roy. Anthrop. Inst., 1917, 

Vol. 47, pp. 37-112. 
Fritsch, G., Zeitschr. f. Ethnol., 1878, Vol. 10, pp. 

15—21; 1880, Vol. 12, pp. 289—300; 1887, Vol. 

19. PP- !95-2°2. 
Frobenius, L., Madzimu Dsangara, Berlin, 1932. 
Gardener, W. E., Scientific American, 1912, Vol. 

107, p. 370. 
Goodall, £., Proc. Rhod. Sci. Ass., 1946, Vol. 41, 

PP- 63-73; Pan Afr. Congr., 1955, pp. 295-9. 
Goodwin, A. J. H., Annals S. Afr. Mus., 1936, Vol. 

24, Pt. 4, pp. 163—210. 
Hall, R. N., Proc. Rhod. Sci. Ass., 1912, Vol. 11, 

pp. 140-154- 
Houghton, S. H., Trans. Roy. Soc. S. Afr., 1926, 

Vol. 13, pp. 105-6; 1927, Vol. 14, p. 315. 
Hewitt, J., Trans. Roy. Soc. S. Afr., 1931, Vol. 19, 

pp. 185—196; Records Albany Museum, 1931, 

Vol. 4, pp. 1—63. 
Halm, T., Zeitschrift f. Ethnol., 1879, Vol. 11, pp. 

307-8. 
Holub, E., Sieben Jahre in Siidafrika, 1880. 
Holm, E., S. Afr. Arch. Bull., 1956, Vol. 11, No. 

41, pp. 12—21; Vol. 13, No. 49, pp. 34 ff.; 

Paideuma, Vol. 6, Sept. 1957, Pt. 5, pp. 297— 

300; IPEK, 19, pp. 77—84; Antaios, I, 5. 
Holm, E., Siidafrikas Urkunst, Pretoria, 1957. 
Hiibner, A., Zeitschrift f. Ethnol., 1871, Vol. 2—3, 

PP- 51-3- 
Impey, S. P., Origin of the Bushmen, Cape Town, 

1926. 
Johnson, J. P., The Prehistoric Period in South 

Africa, London, 1910. 
Jones, N., The Stone Age in Rhodesia, Oxford, 

1926. 



Jones, N., The Prehistory of Southern Rhodesia, 

Cambridge, 1949. 
Leakey, L. S. B., Stone Age Africa, Oxford, 1926. 
Lowe, C. van Riet, S. Afr. Journ. Sci., 1933, Vol. 

30, pp. 525—6; Arch. Series, No. 1, 1938; No. 

5, 1941; S. Afr. Journ. Sci., 1945, Vol. 41, pp. 

329—344; Trans. Roy. Soc. S. Afr., 1937, Vol. 

24, pp. 253—261; Bantu Studies, 1929, Vol. 3, 

pp. 385—393; S. Afr. Arch. Bull., 1947, Vol. 2, 

No. 6, pp. 41—5; Vol. 1, No. 2, pp. 38—40. 
Luschan, F. von, Zeitschrift f. Ethnol., 1908, Vol. 

40, pp. 665—685. 
Mason, A. Y., Bantu Studies, 1933, Vol. 7, pp. 

131-158. 
Moszeik, O., Die Malereien der Buschmanner in 

Siidafrika, Berlin, 1910. 
Obermaier, H. and Kiihn, H., Buschmannkunst 

(English and German), 1930. 
Orpen, J. M., Cape Monthly Mag., N.S., 1874, 

Vol. 9, pp. 1—10. 
Passarge, S., Die Buschmanner der Kalahari, Ber- 
lin, 1907. 
Poch, R., Zeitschrift f. Ethnol., 1910, Vol. 42, pp. 

357-3 61 - 
Prozesky, H., Zeitschrift f. Ethnol., 1906, Vol. 38, 

pp. 908—910. 
Peringuey, L., Trans. Roy. Soc. S. Afr., 1914—5, 

Vol. 4; 1919— 1920, Vol. 8. 
Riet, J. and M. v. d. and Bleek, D. F., More Rock 

Paintings in South Africa, London, 1940. 
Schapera, I., S. Afr. Journ. Sci., 1925, Vol. 22, pp. 

5 4-5i5- 
Stow, G. W. and Bleek, D. F., Rock paintings in 

South Africa, London, 1930. 
Tongue, M. H. and Bleek, D. F., Bushman Paint- 
ings, Oxford, 1909. 
Willcox, A. R., Rock Paintings of the Drakens- 

berg. London. 1956. 
Wilman, M., The Rock Engravings of Griqua- 

land West and Bechuanaland, South Africa, 

Cambridge, Kimberley, 1933. 
Zelizko, J. V., Felsgravierungen der siidafrikani- 

schen Buschmanner, Leipzig, 1925. 

ROCK ART OF AUSTRALIA 

The bibliography relating to this chapter is on 
p. 230. 



236 



PERIODICALS AND ABBREVIATIONS 



Annals S. Aft Mus.: Annals of the South African 

Museum. 
Arch. Series: Archaeological Survey Series. 
Cape Monthly Mag.: Cape Monthly Magazine 

(New Series). 
Bull, de la Soc. preh.: Bulletin de la Societe pre- 

historique. 
IPEK: Jahrbuch fiir Prahistorische und Ethnolo- 

gische Kunst. 
Journ. Roy. Anthrop. Inst.: Journal of the Royal 

Anthropological Institute. 
Mem. Ac. Sciences: Memoires de l'Academie des 

Sciences. 



Pan Afr. Congr.: Proceedings of the Pan-African 

Congress. 
Proc. Rhod. Sci. Ass.: Proceedings of the Rhode- 

sian Scientific Association. 
S. Afr. Arch. Bull.: South African Archaeological 

Bulletin. 
S. Afr. Journ. Sci.: South African Journal of 

Science. 
Trans. Roy. Soc. S. Afr.: Transactions of the Royal 

Society, South Africa. 
Zeitschr. f. Ethnol.: Zeitschrift fiir Ethnologic 



237 



CHRONOLOGY 

Comparative table showing the duration of rock art in the various centres treated in this volume. 
N.B.: The scale changes after 10,000 B.C. and the beginning of the Christian era. 



Middle Ages 

Dark Ages 

Antiquity 
Iron Age 
Bronze Age 
Neolithic 
(Late Stone Age) 

Mesolithic 
(Middle Stone Age) 



Upper Palaeolithic 
(Upper Old Stone Age) 



Lower Palaeolithic 
(Lower Old Stone Age) 



Present 
1800 
1600 

1400 
1200 
1000 
800 
600 
400 
200 



Franco- 
Cantabrian 

Zone 
(Ice Age art) 



Eastern 

Spain 

(Spanish 

Levantine 

art) 



North 
Africa 

and 
Sahara 



South Australia 
Africa 



B.C. 

1000 
2000 
3000 
4000 
5000 
6000 
7000 
8000 
gooo 



10,000 
20,000 
30,000 
40,000 
50,000 
60,000 
70,000 
80,000 
90,000 
100,000 






late offshoots 

probable or 
verified 

possible 



dating 



GLOSSARY 



Acheulean culture 

Culture of the Lower Old Stone Age (Lower 
Palaeolithic), named after the site of St. 
Acheul, near Amiens. The most characteristic 
tool is the hand-axe, a rather crude implement 
made of a flint nodule. Dating: second and 
third interglacial periods (approx. 480,000— 
150,000 B.C.). 

Addax 

Species of African antelope. 

Adrar 

Berber for mountainous region. 

Atlanthropos mauretanicus 

Primitive type of man from the Lower Pleisto- 
cene, of which evidence has been found at 
Palikao, near Ternifine, western Algeria. 

Aurignacian culture 

Earliest stage of the Upper Palaeolithic (Up- 
per Old Stone Age), named after the cave of 
Aurignac (Haute-Garonne Dep.). Advanced 
stage of hunter culture. Blade industry (man- 
ufacture of long flint flakes worked into scrap- 
ers, burins, borers, etc.). Implements made of 
bone, antlers and (mammoth) ivory. Portable 
and mural art. Approx. 60/40,000—20,000 B.C. 
Nowadays Aurignacian is divided into several 
sub-periods: Lower Perigordian or Chatelper- 
ronian, Aurignacian proper, Upper Perigor- 
dian or Gravettian. Main areas of distribution: 
Western Europe, Asia Minor, Russia. Most 
important human type: Cro-Magnon. 

Boskop type 

In South Africa, most important Upper 
Pleistocene representative of homo sapiens 
diluvialis, named after the find of a skeleton 
near Boskop, South-western Transvaal. 

Bovidae 

(From Latin bos). Collective term for wild and 
domesticated cattle. 
Campignian culture 

Late Mesolithic culture in northern France, 
forming a transition to Neolithic culture. 
Named after the hill of Campigny (Seine- 
Inferieure Dep.). 
Canna 

Term applied by Bushmen to a genus of plant 
favoured by elands, and also to this species 
of antelope. 
Capsian culture 

Mesolithic culture in North Africa with origins 



in the Upper Palaeolithic. Named after Capsa 
(now Gafsa) in southern Tunisia. 

Cervidae 

Collective term for all species of deer (incl. 
reindeer). 

Chellean culture 

Earliest stage of the Lower Old Stone Age 
(Lower Palaeolithic), characterized by hand- 
axes. Named after the site of Chelles, near 
Paris. Nowadays the term Abbevillian is 
preferred (after Abbeville on the Somme). 
Dating: first interglacial period (approx. 
540,000—480,000 B.C.). 

Climatic optimum 

Postglacial phase of particularly warm climat- 
ic conditions (approx. 4th and 3rd millennia 
B.C.). 

Cro-Magnon man 

Most important representative of the late 
glacial homo sapiens diluvialis. Named after 
the cave of Cromagnon at Les Eyzies (Dor- 
dogne). Associated with Upper Palaeolithic 
cultures. 

Diabase 

or dolerite: an igneous rock, dark green to 
black in colour. 

Epipalaeolithic 

Transitional period between Palaeolithic and 
Mesolithic. 

Equidae 

Animals of the horse family, including ass, 
zebra and wild ass. 

Equus capensis 

(Cape horse). One of the extinct animal species 
testifying to the antiquity of Bushman art, 
others being the prehistoric gnu, the wart- 
hog, a tapir-like animal and a llama-like 
species. The equus capensis has been described 
by R. Broom on the strength of fossil finds as 
a gigantic horse with a large head and crooked 
nose. 

Fauresmith culture 

In South Africa, Lower Palaeolithic culture of 
the Middle Pleistocene with an improved 
method of producing flakes. Named after the 
site of Fauresmith, in the Orange Free State. 

Felidae 

Animals of the cat family (lion, tiger, etc.). 

Haematite 

or hematite: a mineral containing ferric oxide, 



which can be used as a red paint pigment. 

Homo sapiens 

Collective term for all the races of mankind 
existing at present. Homo sapiens was preceded 
by homo sapiens diluvialis, who succeeded 
Neanderthal man in the Upper Palaeothic. 

Levalloisian culture 

Lower Palaeolithic culture with an improved 
method of producing flakes. Approx. 150,000— 
80,000 B.C., found especially in northern 
France and southern England. Named after 
the site of Levallois-Perret, near Paris. 

Levalloisian technique 

('tortoise core' technique): improved method 
of producing flakes in the Late Lower 
Palaeolithic. The raw material (usually a 
flint nodule) is prepared by blows dealt in a 
certain adroit manner until a relatively large 
flake is struck off. Widely diffused. 

Limonite 

Mineral containing ferric oxide of brownish 
colour, which can be used as a paint pigment. 

Macaroni 

Term for Early Upper Palaeolithic wall 
drawings executed with the fingers or a 
pronged instrument. At first they are merely 
an indecipherable tangle of lines, but later 
they form primitive animal figures. 

Magdalenian culture 

Late Upper Palaeolithic culture named after 
the rock-shelters at La Madeleine (Dordogne). 
Climax of Ice Age reindeer-hunting culture. 
Engravings, polychrome paintings, portable 
art. Approx. 20,000—10,000 B.C. Diffused 
eastwards from Western Europe. 

Maglemosian culture 

Stage of culture in the Mesolithic, named 
after a moor near Miillerup on the western 
coast of Zealand (Denmark), where Mesolithic 
settlements were excavated in 1900. 

Magosian culture 

In East Africa, a culture of the fourth inter- 
pluvial epoch (approx. 10,000—8000 B.C.), 
named after Magosi in Kavamodja. 

Mesolithic 

(Middle Stone Age). Transitional stage from 
Palaeolithic to Neolithic. Dating in Europe 
approx. 10/8000—3000 B.C. (ending earlier in 
the Near East and Egypt). Several sub-groups 
such as Azilian, Sauveterrian, Tardenoisian, 
Maglemosian, Campignian, Capsian (North 
Africa), Natufian (Asia) cultures. The Mes- 
olithic period commences approximately at 



the close of the Pleistocene. Implements (tools 
and weapons) differ only slightly from those 
of the Upper Palaeolithic, but the stone im- 
plements are smaller (microliths, often geo- 
metric in form). Fishing and catching birds 
begin to be important means of livelihood in 
addition to hunting. The dog becomes man's 
first domesticated animal. 

Microlithic artefacts 

(microliths): small stone implements, often geo- 
metric in form (triangles, trapezoids, cres- 
cents, etc.), the appearance of which is 
characteristic of Mesolithic cultures (both in 
the chronological sense and from the stand- 
point of cultural history). Rows of these 
microliths were often inset into shafts of 
bone, antlers or wood. 

Mousterian culture 

Important culture of the Late Lower Pa- 
laeolithic, characterized by an improved 
method of producing flakes (similar to the 
Levalloisian technique) and such types of stone 
implement as flint points and scrapers. It is 
distributed over wide areas of the eastern hem- 
isphere. Mousterian is divided into numerous 
sub-stages and sub-groups, so that nowadays 
it serves only as a collective term. This epoch 
is characterized by Neanderthal man. 

Natufian culture 

In Syria and Palestine, a culture of Mesolithic 
hunters, fishermen and food-gatherers. Ap- 
prox. 8000—6000 B.C. 

Neanderthal man 

(homo neandertalensis). Early man of the 
Upper Pleistocene. Named after the first 
skeleton find of this type in the Neanderthal, 
near Diisseldorf (1856). Remains of Neander- 
thal man are distributed over wide areas of 
the eastern hemisphere. This type of man, 
representative of Lower Palaeolithic cultures 
with an improved method of producing flakes, 
died out with the beginning of the Upper 
Palaeolithic. 

Neolithic 

(New Stone Age). In Europe, dating from 
3000—1800 B.C. (in the eastern Mediterranean 
area beginning as early as 5000 B.C.). First 
elements of a farming economy with fixed 
settlements, keeping of domesticated animals 
(dogs, pigs, sheep, goats, cattle) and cultiva- 
tion of crops. Other innovations in this period 
are pottery, polished stone implements and 
weaving. 



240 



Oryx 

Species of African antelope with horns about 
4 ft. long. ' 

Palaeolithic . 

(Old Stone Age). This is sub-divided into the 
Lower Palaeolithic, approx. 600,000—60/40,000 
B.C., and the Upper Palaeolithic, to 10/8000 
B.C. Order of the individual cultures, which 
often overlap: Abbevillian (formerly Chel- 
lean), Clactonian, Acheulian, Tayacian, Leval- 
loisian, Mousterian, Aurignacian, Solutrian, 
Magdalenian; the last three are Upper Palaeo- 
lithic. The first art finds date from the Aurig- 
nacian (60/40,000—20,000 B.C., the middle of 
the last glacial period). Lower Palaeolithic 
man was a nomadic hunter (food-gatherer, 
hunter, fisherman), but Upper Palaeolithic 
man was a specialized hunter (a so-called 
advanced hunter). Implements were made of 
stone, antlers, bone, ivory and wood. 

Perigordian culture 

Upper and Lower Perigordian are sub-stages 
of Aurignacian. Named after the Perigord 
region in south-western France. 

Perspective, twisted 

With animal bodies depicted in profile, rep- 
resentation of the horns and antlers not viewed 
from the side, but from the front. 

Pleistocene 

(Ice Age or diluvium). Period characterized 
by great climatic changes (Ice Age and inter- 
glacial periods) between the Pliocene (Late 
Tertiary period) and the Holocene (in geolog- 
ical reckoning, the present day), approx. 
600,000—10/8000 B.C. 

Pschent 

Egyptian head-dress. Emblem of royalty. 

Quagga 

(equus quagga): species of zebra which became 
extinct in South Africa in historical times. 

Rock-shelter 

(abri): protected site beneath overhanging rock 
which can serve as a place of settlement. 

Sangoan culture 

In Rhodesia and the Congo, a Middle or Late 
Pleistocene culture of the Lower Palaeolithic, 
having affinities with Fauresmith culture. 

Scalariform signs 

Rock engravings in the form of rows of bands, 
resembling a ladder. 



Solutrian-Magdalenian 

cf. Solutrian and Magdalenian. 

Solutrian 

Upper Palaeolithic culture between Aurigna- 
cian and Magdalenian and differing from these 
cultures in essentials. Characteristic are the 
so-called leaf-shaped spear-heads (feuilles-de- 
laurier) and barbed arrow-heads. Approx. 
30,000—20,000 B.C. Origin still disputed. 
Radiation eastwards from Western Europe. 
Named after the site of Solutre, near 
Macon. 

Stellenbosch culture 

In South Africa, hand-axe industry of the 
Lower Pleistocene, having affinities with Ab- 
bevillian (Chellean). Named after Stellen- 
bosch, about 30 miles from Capetown. 

Tati Bushmen 

Group of Bushmen in north-western Bechua- 
naland and Southern Rhodesia. 

Tectiforms signs 

Rock pictures believed to represent hut- or 
tent-like constructions or traps. 

Tjurunga 

(churinga): longish objects, oval and flat, made 
of stone or wood, usually decorated, used in 
rites by Australian aborigines. 

Totem 

(derived from ototeman, a word used by the 
Ojibway Indians on Lake Superior): a term 
employed by ethnologists to express the idea 
that a mystical connection exists between a 
group of men and an animal, or more rarely 
a plant or natural phenomenon. 

Wild ass 

(equus hemionus): independent sub-species of 
the species equus (other sub-species being the 
horse, ass and zebra). The most important 
breed is the kulan, found in Western Asia. 

Wilton culture 

Group of hunter cultures with microlithic 
stone implements in the northern part of 
South Africa and the Congo. After 8/7000 
B.C.(?). Named after Wilton, in Cape Pro- 
vince. 

Zimbabwe 

Extensive ruins in Southern Rhodesia, dating 
mainly from the 9th— 13th and 17th centuries 
A.D. Their origin and cultural attribution are 
still a mystery. 



241 



INDEX 



The numerals in italics refer to the plates and 
figures. The letter (G) indicates Glossary. 



Abydos 
Adam, L. 
Addaura 
addax (G), 
Adrar (G), 



2IJ 
223 

68, 69, 97 

109, 118 

103 



Adrar des Iforas 103, 118, 779, 131, 138, 139, 142 
Aegadian Is. 68; cf. Levanzo 

Aflou ioi 

Africa, South-West 167, 172, 198 

'Agnus Dei': cf. Pair-non-Pair 
agriculture, agriculturalists 124, 14;, 153, 162, 

163, 203 
Agua Amarga, Cueva del Val del Charco del, 74, 

75, 81, 84 
Ahnet 101, 102, 114 

Ain Gudeja, Mt. 173, 797 

Air 99, 101, 103, 118, 138 

Alac6n 76 

Albarracin 73, 75, 76, 87, 94 

Albert of Monaco, Prince 16 

Alcalde del Rio 16, 63 

Alcaniz 75 

Algeria 99, 135, 173; cf. Oran, southern 

Almagro, M. 85, 86, 90, 91 

Almerian 95 

Alpera 74, 82, 84, 87, pi, 92, 97, 08 

Altamira 15, 16, 20, 21, 24L, 27, 29-31, 32 

63f., 196, 197 
Amazzar, Wadi 124 

Ameib 770, 172 

Amon 138, 150 

Ampoulange, Dr. 57 

Amrah, El 150 

Anahef 101 

Anai 101 

Angles-sur-Anglin 61, 62, 63 

Angouleme 6t 

antelope 20, 91, 99, 109, 115, 116, 118, 156, 

163, 168, 171, 179, 186, 180, 195, ig8f. 
anthropomorphic figures 67, 68, 82, 87, 96, 205, 

209, 21 2f., 217, 22of. 
'Antinea': cf. Jabbaren 
antiquity: of Franco-Cantabrian art 2of., 60; of 

North African art 103-111, 125-133, 136, 142, 

146-152; of South African art 156-9, 162, 165, 

178, 179, 188, 191; of Spanish Levantine art 75, 

88-92; cf. radio-carbon dating 
Aouilalam 779 



Aozon 113 

Arabi, Mt. 75 

Arabia, Arabs 99, 1413, 145; Arab saddle 118, 134, 

145 
Arana, Cuevas de la 75, 82, 86, 87, 02, 04 

Arctic art 13 

Ardales 67 

Ardeche gorge 46 

Ares del Maestre 75, 81 

Arli 142 

Arnheni Land 217L; cf. Oenpelli 

Arquera, Abrigo del 76 

art, Stone Age: cf. Australian art; Franco-Canta- 
brian art; North African art; South African 
art; Spanish Levantine art; antiquity of; pres- 
ervation of; significance of; style; technique 
artefacts (G), i58f., 166, 179 

Artemis 194 

Asbystae 131 

Asia 173 

ass, wild (G), 68, 69, 91, 109 

Atapuerca 66 

Atarantes 131 

Atlantes 131 

Atlanthropos of Palikao (G), 122 

Atlas Mts. 99, 103, 1S3 

Auanrhet, 'White Lady' of 136, 141, 144 

Augila 131 

Aurignacian (G), 16, 20, 21, 24, 33, 35, 43f., 4;, 
5 J . 5 1 . 54^-. 6°> 64, 67, 190, 199; Aurignacian- 
Perigordian 3, 20, 21, 33, 37, 38, 41, 47, 54, 58, 
61,65 
aurochs 26, 30, 33, 38, 39, 46, 60, 66, 69, 73 

Australian art 205-229; extraneous influences on 
216-7, 228; Central Australia 212, 224f., 227, 
229; mid-west 208; northern 208, 217-221, 206, 
277-3, 229; north-western 205-217, 207, 210, 214, 
277, 210; south-eastern 208, 221, 227, 228; 
southern 223, 224-7, 22 9- south-western 208; 
western 220, 224-6; cf. New South Wales; 
Queensland; Victoria 
Aventignan 50 

Ayers Rock 224, 227 



Bafour 

Bailloud, M. 

Bambata 

Bantu 

Barabao 

Bardai 

bark painting 



140 
102 

179 

153, 162, 202 

57.58 

707, 11 j, 720 

220, 221, 225, 226, 229 



242 



Barth, H. 
Bar)', E. v. 
Basedow, H. 
Baume-Latrone 
Bayol, Abbe 



99- "» 
101 
218 

45. 45> 6 °. 67. !7 2 
46 



bear 28, 331., 44, 47, 49, 50, 53, 58, 61, 64, 91 

B^deilhac 25 

Begouen, Count 16,43,48 

Beni-Ounif 101 

Bernifal 16 

Bes Seba, Djebel 173, 183 

Bessac, M. 62 

Bethlehem 197 

Bicorp 75, 82, 87 

birds 49, 53, 92, 206, 222; eagle 270; eagle-hawk 

21 1; ostrich 99, 104, 109, 115, 118, 152, 153, 169, 

222; owl 44, 46, 142; pelican 112 
bison 20, 26, 27, 2gf., 30, 31, 34, 37, 40, 40-2, 43f., 

47 > 5°. 53- 66 . 62, 173, 196 
Black, L. 222 

Blanc. Baron G. A. 68 

Blanchard 55 

Bleek, Miss D. 171 

Bleek, W. 167, 200 

Blunder Bay 218 

boar 30, 68, 84, 92 

boat 139, 148 

Boskop man (G), 165, 166 

Bou Alem 101 

Bourg-sur-Gironde 60 

bovidae (G), 50, 55, 87, 99 

Bradshaw, J. 214, 216; cf. style, 'elegant' 

Brandberg, 'White Lady' of 149 

Brenans 102 

Breuil, H. 15, 16, 20, 21, 24, 33f„ 44, 46, 50, 53!., 

731., 88, 90, 127, 146, 162, 172; Breuil gallery 

47, 47, 48, 59, 60, 69 
bubalus 111, 115, 135, 150; bubalus artist 134, 135; 

bubalus period 104, 107, no, 110, 112, 114, 114, 

115, 119, 125, 127, 128, 132, 133, 136, 142, 146-7, 

150, 152 
buffalo 103, 109, 122; cf. bubalus 

Bugue, Le 57, 5 8 

Bulawayo 175 

burial 166, 167 

Burra 224 

Bushmen 12, 98, 135, 146, 149, 1531., i64f., 171, 

2 73' 1 78. 183, 186, ig4f.; Tati Bushmen (G) 
Buxu 65, 66 



Caballos, Cueva de los 


75, 80, 81, 81 


Cabre, J. 


66,73 


Cabreret 


53 



Calapata 73 

camel 101, 103, 105, io8f., no, 116, 130, 132, 134, 

■ 145; camel period 113, 114, 118, ng, 132, 133, 
145; cf. riding 

Campignian (G), 128 

canna (G), 194 

Cap Blanc 25, 56, 57, 57, 61 

Cape Province 166, 167, 198 

Capitan, L. 16 

Capsian (G), 94 f., 135 

Cardes 65 

Carson River 213 

Cartailhac, E. 16 

Carthage 99, 101 

Casares, Los 66, 67 

Castanet 55 

Casteret, N. 49, 50 

Castillo, El 16,20,21,63^ 

Castro-Marino 68 
cattle 34, 54, 58, 64, 65, 99, ioif., io8f., 115, 118, 

128, 130, 133, 136, 13S, 139, 141, 152, 202; cows 

37, 39, 58, 68, 196; oxen 3, 32, 39, 47, 53, 58, 68, 

74< 75- 79' 8 5< 8 7> 92. 97- IoS - 10 9> JI °- »5» II6 - 

117, 120, 135, 148, 174, ig6, 200; cattle-raising 

101, 124, 125, 140, 153, 200 
Cavaignac, Gen. 99 

cave-bear: cf. bear 
celestial bodies 163, i68f., 174, i78f., 196, 202; 

cf. Mantis 
Cerro Felio 76 

cervidae (G), 33, 65, 69, 96 

Chabot 46 

Chad 140 

Chaire a Calvin 61 

chamois 92 

chariot 103, 109, 113, 116, 129, 131, 142-3, 146; 

cf. war chariot 
Chebkha Dirhem 101 

Chelles-Acheul (G), 122, 158 

Chiron, L. 46 

Civil, Cueva del 75, 76, 79, 83 

clay 25, 26, 43, 48, 49, 52f., 60, 62 

climate 69, 77, 89, 95, 101, 109, i22f., 132; 

climatic optimum (G), 95 
clothing 80, 82, 84, 88, 94, 93, 95, 97, 126, 137, 

140, 143, 149, 214 
Cocinilla del Obispo g4 

Cogul 74 79, 82f., 94, g7, 97, 149 

Colombier 46 

Combarelles, Les 16,20,33^,55,36 

Conic Range 223, 224, 22S 

Covalanas 16, 21, 63, 66, 67 

cow: cf. cattle 



243 



95- 


129, 


M9 

124 
222 




129. 


131 



Crete 
crocodile 

Cro-Magnon man (G), 
cults: cf. religion 
culture hero 
Cyrenaica 

Daleau, F. 15, 60 

Dalloni 102 

dance 81, 82, 84, 8y, 141, 154, 195, 196; 'Dancing 

Women': cf. Cogul 
Darfur 145 

David, A. 53 

Davidson, D. S. 214,218,222,223 

deer 92; hind 29, 30, 37, 32, 34, 63f., 67, 68, 86; 

stag 30, 34, 38, 39, 39, 40, 43, 44, 53, 55, 58, 61, 

64f., 73, 75, 81, 86, 87, 92, 96, 98; cf. antelope; 

chamois; elk 
Delamere 218,224,225 

demon 169, 174, 182, 183, 206, 214, 218 

Depuch I. 220 

Devon Downs 224, 229, 229 

diabase (G), 755, 160, 164, 181, 184, 188, 793 

dingo 2o6 

Djerat, Wadi 102, 104, 109, 770, 111, 772, 774, 

115, 726, 127, 730, 134, 150 
dog 92, 98, 115; cf. dingo 

Dogues, Les 76, 81 

Dokou 101 

Dona Clotilde, Cueva de 76, 94 

Drakensberg Mts. 162, 163, 167, 171, 172, 772, 

i94f., 797, 202 
Drysdale River 213 

Duveyrier, H. 101 

Ebbou 46 

Egypt 99 f - I26 - 12 9. 1 3 e > !3 8 . I 4 I > 1 i3> '45> 

i4i8f., 757, 200 
eland 763, 765, 170, 772, 173, 174, 186, i92f., 

793, 797, 202 
elephant 45, 65, 66, 99, 103, 707, io8f., 115, 122, 

139, 142, 150, 152, 766, 171 f., 200 
elk 30, 92 

Els Secans 75, 84, 9; 

Ennedi 99, 102, 119, 138 

equidae (G), 40, 69 

equus capensis (G), 190 

equus hemionus: cf. ass, wild 
erg 122, 732 

Erongo Mts. 770 

Ethiopians 139, 140 

Eucolo Creek 224 

Euriowie 225, 226 

Eyzies, Les 16, 32, 33, 56 



fables 167, 168, 171, 183, 188, 198-9 

Fauresmith (G), 158 

felidae (G), 33, 35, 60, 61, 68 

fertility 35, 36, 44, 48, 175, 196, 200 

fetish 99 

Fezzan 99, 101, 102, 109, 113, 131, 134, 135, 143, 

150, 152 
Figuier, Le 46 

fish 122, 124, 206, 206, 218, 222; pike 55; trout 

43, 44; tunny-fish 65; fishing 122, 153 
Flamand, G. B. M. 113,114 

Flinders Ranges 224 

'flying gallop' 113, 129, 729, 131, 142, 143 

Font-de-Gaume 16, 21, 32, 33, $4, 196 

food-gatherers 122, 153, 154 

foreshortening 173, 193, 196, 797 

Forest River 218 

Foum el Hassane 103 

Foureau, F. 101 

fox 34 

Franco-Cantabrian art 15-72; affinities with North 

African art 145-9; with South African art 170- 

3, 197; with Spanish Levantine art 6g, 73, 90-1, 

96-7 

Frobenius, L. 101, 113, 149, 175, 182, 192 

Fula 140, 142. !4 8 

Futa Jallon 140 



Ganties 
Gao 

Garamantes 
Gargas 
Garrigou, Dr. 
Garrod, D. 



48,49 

142 

131, 142, 143 

16, 20, 21, 5of., 57 

41 

62 



Gasulla gorge 75, 76, 81, 82, 85, 87, Sy, 90, 93, 96 
gazelle 99, 109, 118, 139 

Gerzeh, El 150 

Gibaja 63 

Gillen 212 

giraffe 111, 115, 118, 723, 139, 142, 152, 769, 

i76f., 181, 182, 7S5, 7*7, 191, iggf. 
giro giro 216 

Glen Isla 222, 223 

Glenelg River 214 

Glory, Abbe 46. 5 8 

glutton 66 

goat 75,81,56,92,99,118,128,138 

Godin, H. 49 

Gosford 221 

Greiribat, El 723 

Greze, La 24. 57 

Groote Eylandt 200, 221 

Grey, G. 214, 217 



244 



haematite (G), 76 

hair-dress 82, 214 

Hamraamat, Wadi 150 

hands, paintings of 21, 29, 41, 45, 46, 511., 57, 

54, 6 3 f. 

Hathor 200 

Haut In Daladj 146 

Hawker 223 

Haza, La 16, 63 
head-dress 8o, 84, 116, 118, 126, 136, 140, 143, 757 

Herberts, K. 76 

Herenveen 772 

Herodotus 129, 131, 132 
hind: cf. deer 
hippopotamus 109, 111, 115, 122, 124, 139, 142, 

•52 
Hoggar ioif., 702, 70S, 114, 116, 118, 119, 131, 134, 

138,73s, 139,145,746,152 
homo sapiens (G), 26 

horse 3, 28, 301., 36, 38, 39, 40, 40, 43-50, 531., 54, 

57> 75> 9 2 > 101 > I02 > I0 3' 10 5> 10 9 r > M 8. is8f., 

134, 143, 145, 190, 200; cf. Arab saddle; equus 

capensis; riding 
horse period 114-118, 726, 129, 130, 730, 133, 143, 

149 
Hottentot 153, 199-200 

Hountao River 49 

human figures 34, 35, 44, 53, 56, 62, 68, 69, 741., 

76, 77, 79, 81, 83-;, 87-92, 95, 95, 97, 97, 991., 

702, 770, HO, 113, 773, ll6, 726, 129, 732, 133f., 

139, 140, 141, 143, i46f., 148, 151, 75S, 763, 769, 

195, I98, 203, 277, 217f., 220, 224, 225, 226, 226, 

22S; grotesque 47, 206, 214; Alpera type 77, 79; 

cestosomatic type 79, 79; nematomorphous type 

79-80, S7; pachypodous type 79, 80 

Humbert River 218 

hunters, hunting 11, 12, 26, 37, 43, 44, 49, 50, 74, 

8of., 80, 8;, 95, 98, 111, 122, 124, 139, 142, 153, 

154, 162, 164, 174, 175, 192, 194, 198, 200, 202f.; 

hunter artists 79, 85, 90, 94, 162, 165, 179, 180, 

195, 203; hunter culture 12, 13, 28, 52, 92, 97, 

98, 122, 162, 203; hunting scenes 84-5, 88, 147, 

175, 221, 227; hunter period: cf. bubalus period 

Hyksos 129 

ibex 30, 33, 34, 43f., 43, 53, 5gf., 61, 81, 85 

Ice Age 13, 15, 19, 20, 24f., 37, 38, 46, 52, 56, 58, 

65, 68f., 73, 88f., 122, 158 
idols 139, 142, 187, 188 

implements 25-6, 43, 56, 94, 105, 106, 121, 125, 
149, i54f., 167, 222, 225; axes 105, 106, 110, 115, 
i2if., 273; flake and pebble tools i2if., 127, 
i57f.; grindstones 124, 128; scrapers 121; thread- 
cross 225 



Inagurdurwil 272 

initiation 48, 161, 192, 196, 200 

In-.Habeter 150 

insects 75, 82, 87, 92, 92-3, 198; cf. Mantis 

instruments, musical 44 

Isis 136, 745 

Isolane 101 

Isturitz 53 

Italy 68-70, 101 



Jabbaren 
jackal 
Janus head 
Jeannel, Dr. 
Juba, King 
justice 

Kalahari 
Kandjanlgari 
kangaroo 
Kargour Thai 
Katherine River 
Kef Bou-Bekeur 
Kemal el Dine, Prince 



136,747,757 
730 

139 

16,46 

132 



156, 163 
211 

206, 222 

'33 
218, 220 
132 
102 
Kimberleys 218, 221, 225f.; eastern 214; western 

214, 216 
King Leopold Ranges 221 

Ksour 99 

kudu 7S5, 207, 203 

Labattut 55, 56, 96 

lalai 209 

Langi Ghiran, Mt. 223 

Lascaux 3, 16, 21, 24, 28, 36f., 37-9, 47, 55, 96, 

183, 184 
Laussel 56, 62; Venus of 56 

Leeufontein 77S, 182, 184 

Lemozi, Abbe 53 

Lentz, Dr. O. . 103 

Levalloisian (G), 158; Levalloisian-Mousterian 

122, 127 
Levanzo 68, 69, 69, 197 

Libya, Libyans 726, 129, 131, 138, 143, 150; cf. 

Cyrenaica, Fezzan 
Libyco-Berber writing 702, 103, 116,730, 143, 145 
limonite (G), 76 

Lindner, K. 82 

lion 28, 34, 3s, 40, 44, 50, 66, 91, 99, 109, 774, 730, 

135, 77S, 182 
Liverpool River 225 

lizard 223 

Llort, Barranco de 76 

Lotophagi 131 



245 



Macae 

'macaroni' (G) 
Madeleine, La (Dordogne), 
Magdelaine, La (Tarn) 



131 

21,52,53,60 

21, 173, 176 

62, 63 



Magdalenian (G), 20, 24, 27, 32!., 34, 41, 42, 43, 

44, 47, 471., 53f., 57, 62, 69, 197, 199 
Maghreb 99 

magic 28, 35:., 43, 49, 50, 85, 88, 149, 160, 161, 

165, 173, 184, 201; magician figure: cf. Trois 

Freres 
Maglemosian (G), 92 

Magosian (G), 159, 198 

Mairie, La 58 

Malkaia 224, 229 

mammoth 20, 26, 32f., 44, 46, 53^, 59!:., 65, 172, 

173 
Mantis 168 f., 179, 183, 186, 194, 195, ig8f. 

Marconi, B. 68 

Mardoch6 ben Serour, Rabbi 103 

Maretjiesfontein 160, 189 

Marrakesh 103 

Martin, H. 61 

Mas d'en Josep, Cueva 75, 84, 94 

mask 35, 69, S7, 96, 1 19, 757, 141, 169 

Matthew 223 

Mathurin, S. 62 

Matopos 169, 174, 175, 185, 1S7, 198, 203 

Mauritania 103, 116, 118, 119, 138, 140, 143 

Mazaleon 75 

McCarthy, F. D. 221, 222 

McDonnel Ranges 225 

Melones, Cueva de los 94 

Mesolithic (G), 90, g2f., 146, 159 

microliths (G); cf. artefacts 

Milton 222 

Minateda 75, 82, 88, 91 

minor arts 20, 53, 64, 97 

Moghar gg, 101 

Molard 16, 43 

Monod, Th. ioif„ 114 

Montauban 62 

Montespan 25, 48f., 49 

Montesquieu-Avantes 16, 43, 48 

Montignac 3 6, 55 

Moors 145 

Mootwingee 224 

Morella la Vella 75,75,81,57 

Morgan, J. 213 

Morocco 99, 103, 134, 138, 143; cf. Moors 

Moros, Roca dels 84, g4, g7 

Mortaja, Barranco de la 75 

Mortero, El 76 

Mouchgueug 101 



moufflon 
Mountford, C. P. 
Mousterian (G), 
Mouthe, La 
mule 
musk-ox 
Mycenae 



105, log, 115, 118, 130 

218 

25, 122, 127, 158 

15, 25, 60 

49 

20, 28, 54, 61 

i2g 



myths, mythology 88, g3, i54f., 162, 167, ijo, 

i7of.. 176, 179, 182L, 18}, 191, ig2f., 202, 205^, 
213, 216, 218, 222, 224, 228; mythical creatures 
38, 3g, 212 



Nachtigal, Dr. 
Narr, K. J. 
Nasamones 

Natufian (G), 
Neanderthal man (G), 



101 
97.98 
i3' 
92 
165 



Negroes, Negro art ng, 121, 136, 139, 147, 149 
Neolithic (G), 90, 94, 95, 106, 108, 110, i2if., 

146, 150, 159 
New South Wales 221-2, 224, 227, 228 

Ngungunda 216, 2/9, 225 

Niaux 16,25,^0-3,41,43,^5,55 

Nicolas, M. 103 

Niger 124, 131, 142 

Nile 138, 148, 150 

Nimes 45, 46 

Niscemie 69 

North African art 99-152; affinities with Cretan art 

i2g; with Egyptian art 101, 136, 150-2; with 

South African art i4g, 173 
Nougier, L. R. 58 

Nswatugi 174, 176-g, 185, 185, 187, 195, igg, 201, 

202, 203 

Obermaier, H. 16, 63, 65, 75, 76, 81-2, 88, go 

Oenpelli 206, 211-3, 217, 218, 220, 221, 228 

Olivonas 75 

Ooraminna 225 



opossum 



206 



Oran, southern ggf., io8f., 113, 115, 125, i32f., 

138, 143, 146-7, 150, 152 
ornaments 80, 82, 84, 88, 128, 135, 137, 141, 

144, 154 
Orpen ig4 

oryx (G), 109, 118 

Ouan-Sidi 132 

Ouenat, Djebel 102, ng, 133, 138 

Oullins 4 6 

ox: cf. cattle 



pachyderm 
Pair-non-Pair 



109, 115, 182 
15, 19, 20, 24, 60, 61 



246 



Palaeolithic (G), 33, 63, 69, 90, 121, 122, 125, 127, 

139, i56f., ,171; Epipalaeolithic (G), 94f.; Lower 

25, 122, 153; Middle 127, 158, 159, 179, 198; 

Upper 21; 33, 46, 53, 55, 56, 63, 90, 91, 94, 96, 

97, 122, 125, 146, 1531., 1581. 

panther 109 

Para River 224 

Parpallo 91, 97 

Pasiega, La 16,21,65,66 

Passemard, E. 53 

pastoralists 128, 1381., 147, 149, 200, 202; pastora- 

list period 114, 115, 117, 119, 120, 123, 129, 131, 

138, 139, 141, 142, 145, 147L, 148, 151, 162, 163 

patina i05f., 111, 114, 115, 116, 118, 142, 178, 

182, 190 
Pech-Merle 21, 47, 53f., 54 

Pellegrino, Mt. 68, 6g, 97 

Pefia de Candamo 66 

Penches 66 

'people of the sea' 129, 131 

Perigordian (G), 20, 21, 24, 25, 43, 51 f., 60, 62, 

63, 66, 67, 68, 90 
perspective, natural 43, 49, 58, 69, 139, 190; twisted 

(G), 3, 24, 25, 38, 351, 41, 46, 55, 66, 172, 193 
Petrie 150 

Peuhl: cf. Fula 

Peyrony, D. 16, 58 

Philipp Cave 170, 172 

Pigeon Hole 218 

Pilangura 2/7 

Pileta, La 17, 45, 67, 68, 90 

Pimba 224 

Pimiango 65 

Pindal 65, 66 

Plassard 58 

Pleistocene (G) 

Polvorin, Cueva del 76 

Pomongwe 198, igg 

Pont-du-Gard 46 

Porcar, J. B. 76 

Portel, Le 16, 21, 46L, 47 

Prado del Navazo g4 

preservation of art works: Australian 216, 224; 
Franco-Cantabrian 17, 19, 50; North African 
108; Spanish Levant 77-9; restoration 87 
Pretoria 755, 165, 180, 181, 192 

pschent (G), 151 

Puente-Viesgo 63, 65 

Puigaudeau, Mile, du 103 

Pygmies 175, 192 



quagga (G), 

Quanciqutschaa 

Queensland 



160, 187, li 



8f., ig8f. 

194 

221, 227 



Rabosa, Cueva de la g4 

radio-carbon dating 128, i2g, 147-8, 224 

Rarmales 63 

Regent River 214 

Regnault, F. 16, 47, 50 

reindeer 20, 28, 32f., 34, 44, 46, 48, 53, 58, gi, 196 
religion, cults 135, 136, 145, 147, 149, 150, 154^, 

177, 179, 186, 192, 194, 198, 212, 228; cf. idols; 

magic; rites 
Remigia, Cingle de la Mola 75-6, 81, 82, 87, 87, 

S9. 9h 9 6 
Remigia, Cueva 75-6,81,(55,90,96 

Reverdit 56 

rhinoceros 26, 33, 37, 45, 58f., 66, 91, 103, 104, 

108, 109, 1 1 1, 1 15, 139, 142, 150-2, 155, 178, l82f., 

19' 

Rhodesia 167, 175, ig6f., 202, 203 

Rhotert, H. 138 

Riba del Saelices 66 

Richa, El 101 
rider, riding 113, 116, 131, 132, 134, 143, 145 

Rio de Oro 103,138 

rites 28, 3sf., 43, 44, 48, 50, 52, 67, 141, 161, 162, 

167, 175, 179, 182, ig2f., 203, 206, 212, 222; cf. 

fertility; initiation 

Riviere, E. 15 

Robert, R. 58 

Roc de Sers, Le 24, 61, 61, 62 

Rodd, Fr. 103 

Rohlfs, G. 108, 109 

Rojals 76 

Romanelli 68 

Roper River 220 

Rouffignac 16,58^,70 
'round-headed men': cf. style 

Sahara ggf., 108, 113, 116, 118, I22f., 131, 138L, 
I 4 2f. 

St. Bertrand de Comminges 50 

St. Martin d'Ardeche 46 

St. Paul, Mt. 166, 171, 172 

St. Perrier, Comte de 53 

Saltadora, Cueva 75, 77, 81, 88 

San 166 

Sangoan (G), 158 

Santander 28, 63 

Santian 63, 66 

Santillana del Mar 15, 28 

Sautuola, Marcelino de 15; Maria de 15, 29, 160 
Schulz, Mrs. A. S. 216, 223 

sculpture 24, 25, 30, 43, 48f., 55f., 61, 61, 62, 108, 

124, 132, 138-9, 139, 143, 15S, 159, 180, 190, ig2, 

205 
Sefar 137, 141, 148 



247 



io3 

• 55 

6 7 

i 5 o, 



Senones, Mile. M. 

Sergeac 20, 2 

Serrania de Ronda 

sheep 128, 138; ram 109, 115, 135, 139, 143, 

152; wether 124, 138; shepherd 148 

Sierra, Padre 16, 63 

significance of art works: Australian 222; Franco- 

Cantabrian 25, 26, 28, 35, 36, 43, 418; North 

African 135; South African 161, 165, 167, 173, 

i84f., 188, 194, 200, 202; Spanish Levantine 84f. 

signs, symbols: club-shaped 48, 63, 65; lattice-like 

39, 40; scalariform (G), 29, 68; trident-shaped 

63; cf. tectiforms 

Silet 138 

Silozwane 169, 203 

Silurian 122 

Siouah 150 

Sireuil 158, 170 

snake 45, 60, 203, 209, 270, 217, 218, 222, 228 

Solutrian (G), 20, 24, 66; Upper 41, 53, 61; 

Solutrian-Magdalenian (G), 20, 24, 25 
South African art 153-203 

Spanish Levantine art 73-98; affinities with Cretan 
art 95; with North African art 94-5, 98, 145-9; 
with South African art 97-8, 197, 200 
Spencer 212 

spirit 44, 87, 96, 212, 218 

springbok i7if., 200 

stag: cf. deer 

Stellenbosch (G), 158 

Strabo 129, 

Strehlow 

style: abstract 91, C)6f., 221, 225, 226, 229; archaic 
45 r > 55> 66; double triangle 702, 11 _j, 113, 116, 
118, 143; 'elegant' (Bradshaw) 214, 2141., 221, 
224, 227, 228; Expressionist 80; geometric 150, 
208, 225, 226; linear 43, 45, 46, 118, 208, 221; 
naturalistic 24, 25, 30, 55, 58, 62, 68, 69, 79, 
inf., 139, 146, 150, 206, 208, 216, 221, 226, 227; 
'round-headed men' 128, 133, 136, 757, 141, 
14;, 147, 149, 757; schematic 24, 79, 80, 82, 1 i2f., 
/20, 142, 143, 146, 216, 229; semi-naturalistic 
68, 104, 1151., 150; silhouette 41, 51, 57, 54-5, 
755, 166, 171, 172, 77^, 176, 177, 184, 189, 198; 
wondjina 2i4f., 227; X-ray 206, 218, 228; cf. 
'flying gallop'; foreshortening; perspective 
Sudan 99, 102, 103, 148; Sudanese steppe 138, 

140, 142 
Sydney 227, 221, 222, 228 



132 
212 



Tabelbalet 

Tafilalet 

Tahtani 



Talaat-n-Iisk 103 

Tamentit 143 

Tanezrouft 124, 131, 142 

Tapsus 132 

Tarascon-sur-Ariege 41 

Tarzerouc 108 

Tassili 101, 103, 104, 109, 770, 772, 774, 119, 724, 
126, 127, 129, 750, 131 f., 134, 737, 747, 745, i46f. 
148, i;i, 152; Tassili-n-Ajjer 102, 119, 136, 142 
technique: Australian 206, 207, 211; Franco-Canta- 
brian 2gf., 4of., 52, 63, 66; North African 103^, 
109, 112, 115, 116, 118, 132, 150, 152; South 
African 156, 180, 184, i8gf., 196, 197, 199; Spa- 
nish Levantine 76, 77 



tectiforms (G), 

Teda 

Tefedest 

Tel Issaghan 

Tenere 

Teyjat 

Thyout 

Ti-m-Missao 

Tiaret 

Tibbu 

Tibesti 99, 1 

145, 152 
tiger 

Tindale, N. B. 
Tit 

tjurunga (G), 
Torm6n 
totemism (G), 



24. 33. 6 5- 66 

138 

99 

124 

16, 58, 68 

99, 101 

131,1412 



131, 145 



01, 102, 113, 116, 119, 131, 13 



28 
220 

702, ll6 
212, 226 

75. 77- 77. 79. 8 7 
88, 173, 212 

Transvaal: eastern 197; northern 198; western 755, 
160, ij8, 1S1, 182, 189, 193 



Troglodytes 

Trois Freres, Les 

Tuareg 

Tuat 

Tuc d'Audoubert 

Tugby, D. J. 

Tunisia 

turtle 

Unambal 
Unbalania 
Ungud 
unicorn 



J3 1 

16, 25, 43, 44, 50, 168, 176 

99, 101, 116, 118, 139, 143 

108, 139, 14 j 

16, 25, 43, 48, 50, 62 

224 

99. *35 

218 



207 

206, 277, 27_J 

2o8f„ 217 



142 
108 

99, 101 



Valltorta gorge 75, 76, 76, 77, 79-S7, 81, 84, 88, 94, 

94 

Varaignes 58 

Varilhes 46 

Vaufrey, Prof. 127 



'Venus': cf. Laussel 
Verner, W. , 
V^zere 

V^zian, M. - 
Victoria 
Victoria River 
Vieja, Cueva 
Villar del Huomo 
Visera, Cantos de la 
Volp River 
Vryburg 



67 

32. 36 
46 

222f., 228, 22<S 
2l8 



75, 82, 84, 87, 



91, 92, 98 
92 

75 

48 

765, 170 



Walanganda 209 

Walgarock 222 

war: chariots 101, 102, 129, 131, 142; scenes 81, 85, 

S-], 142-3, 147, 162; warriors 76, 81-2, 8y, 88, 126, 

134, 272, 273 

weapons 49, 80, 82, 88, 143, 145, 159, 161, 179, 

207; arrow 48, 54, 56, 7./, 81, 82, 96, 161, 162; 

arrow-head 121, 124, 128, 159, 167; boomerang 

105, 110, 115, 222, 223; bow 56, 7./, 82, 91, 9J, 

9g, 105, 110, 115, 116; club 115; fire-arms 105, 

110, 118, 203; javelin 82, 105, 110, 116, 118, i2g, 

730, 73,/, 143, 769, 77./, 194, 202; knife 116, 143, 



145; lance 110; lasso 82; lasso-thrower 94; shield 

110, 116, 129, 730, ijj, 143, 162, 222; spear 50, 

•272, 218; spear-thrower 272, 218, 220; sword 

105, 110, 118 

Wernert, P. 16, 63, 65, 79 

'White Lady': cf. Auanrhet, Brandberg 

Wilton (G), 166 

Windels, F. 16,58 

Winkler, H. A. 133, 150 

wolf 30L, 66 

Wonalirri 27./ 

woyidjina 205f., 207, 2iof., 228; cf. style 

Worms, E. A. 216 

writing 21, 101, /02, 103, 116, 730, 143, 145 

Yule River 221, 225 



zebra 
zebu 

Zeramour 
Zimbabwe (G), 
Zohrer, M. 
zoomorphic figures 
228 



122, 77./, 7S5, 188, if 



212, 213, 216, 220, 221 



118 
103 
203 
103 
226, 



249 



<h 



659F14^ 

84/13/39 347E5 



MARSTON SCIENCE LIBRARY 
Date Due 



Due 


Returned 


Due 


Returned 



















































































































































































3 1262 07005 6261